4.6_RF.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import packages\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import math\n",
    "import sklearn\n",
    "import sklearn.preprocessing\n",
    "import datetime\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.model_selection import ParameterGrid\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=Warning)\n",
    "import seaborn as sns\n",
    "sns.set()\n",
    "pd.options.mode.chained_assignment = None  # default='warn'\n",
    "\n",
    "#Program that executes the rolling window train-val-test split \n",
    "%run TimeBasedCV.ipynb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>eom</th>\n",
       "      <th>prc</th>\n",
       "      <th>me</th>\n",
       "      <th>ret</th>\n",
       "      <th>ret_exc</th>\n",
       "      <th>ret_exc_lead1m</th>\n",
       "      <th>ret_6_1</th>\n",
       "      <th>ret_12_1</th>\n",
       "      <th>tax_gr1a</th>\n",
       "      <th>...</th>\n",
       "      <th>cowc_gr1a</th>\n",
       "      <th>pi_nix</th>\n",
       "      <th>ret_6_0</th>\n",
       "      <th>ret_1_0</th>\n",
       "      <th>noa_at</th>\n",
       "      <th>cop_atl1</th>\n",
       "      <th>prc_highprc_252d</th>\n",
       "      <th>ocf_at</th>\n",
       "      <th>dbnetis_at</th>\n",
       "      <th>netdebt_me</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>comp_001166_02W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>11.486148</td>\n",
       "      <td>567.074646</td>\n",
       "      <td>-0.018948</td>\n",
       "      <td>-0.019948</td>\n",
       "      <td>-0.089504</td>\n",
       "      <td>-0.229560</td>\n",
       "      <td>-0.353251</td>\n",
       "      <td>-0.020556</td>\n",
       "      <td>...</td>\n",
       "      <td>0.041844</td>\n",
       "      <td>3.815513</td>\n",
       "      <td>-0.244158</td>\n",
       "      <td>-0.018948</td>\n",
       "      <td>0.692706</td>\n",
       "      <td>0.093015</td>\n",
       "      <td>0.398130</td>\n",
       "      <td>-0.007700</td>\n",
       "      <td>0.085202</td>\n",
       "      <td>0.135368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>comp_001661_01W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>35.853958</td>\n",
       "      <td>5221.931641</td>\n",
       "      <td>0.005815</td>\n",
       "      <td>0.004815</td>\n",
       "      <td>0.108428</td>\n",
       "      <td>0.129149</td>\n",
       "      <td>-0.035109</td>\n",
       "      <td>-0.033306</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.019596</td>\n",
       "      <td>1.284930</td>\n",
       "      <td>0.135715</td>\n",
       "      <td>0.005815</td>\n",
       "      <td>0.751569</td>\n",
       "      <td>0.164124</td>\n",
       "      <td>0.918059</td>\n",
       "      <td>0.096164</td>\n",
       "      <td>0.063714</td>\n",
       "      <td>0.200036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>comp_004367_02W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>36.068653</td>\n",
       "      <td>4139.599121</td>\n",
       "      <td>-0.103801</td>\n",
       "      <td>-0.104801</td>\n",
       "      <td>0.127227</td>\n",
       "      <td>-0.020286</td>\n",
       "      <td>0.087151</td>\n",
       "      <td>-0.037240</td>\n",
       "      <td>...</td>\n",
       "      <td>0.026114</td>\n",
       "      <td>1.725004</td>\n",
       "      <td>-0.121981</td>\n",
       "      <td>-0.103801</td>\n",
       "      <td>0.961141</td>\n",
       "      <td>0.240563</td>\n",
       "      <td>0.675468</td>\n",
       "      <td>0.070724</td>\n",
       "      <td>-0.000607</td>\n",
       "      <td>0.410184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>comp_004925_02W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>29.735168</td>\n",
       "      <td>15302.482422</td>\n",
       "      <td>-0.070425</td>\n",
       "      <td>-0.071425</td>\n",
       "      <td>0.064832</td>\n",
       "      <td>0.046064</td>\n",
       "      <td>0.023265</td>\n",
       "      <td>-0.007344</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.000724</td>\n",
       "      <td>1.961724</td>\n",
       "      <td>-0.027605</td>\n",
       "      <td>-0.070425</td>\n",
       "      <td>0.466852</td>\n",
       "      <td>0.221983</td>\n",
       "      <td>0.861333</td>\n",
       "      <td>0.084234</td>\n",
       "      <td>-0.008142</td>\n",
       "      <td>-0.075232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>comp_005959_01W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>39.181721</td>\n",
       "      <td>6355.691406</td>\n",
       "      <td>-0.085813</td>\n",
       "      <td>-0.086813</td>\n",
       "      <td>0.001832</td>\n",
       "      <td>0.123795</td>\n",
       "      <td>0.012811</td>\n",
       "      <td>-0.012241</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.014214</td>\n",
       "      <td>1.069862</td>\n",
       "      <td>0.027359</td>\n",
       "      <td>-0.085813</td>\n",
       "      <td>0.635742</td>\n",
       "      <td>0.108581</td>\n",
       "      <td>0.721256</td>\n",
       "      <td>0.069331</td>\n",
       "      <td>-0.038594</td>\n",
       "      <td>0.500641</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 55 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                id        eom        prc            me       ret   ret_exc  \\\n",
       "0  comp_001166_02W 2003-01-31  11.486148    567.074646 -0.018948 -0.019948   \n",
       "1  comp_001661_01W 2003-01-31  35.853958   5221.931641  0.005815  0.004815   \n",
       "2  comp_004367_02W 2003-01-31  36.068653   4139.599121 -0.103801 -0.104801   \n",
       "3  comp_004925_02W 2003-01-31  29.735168  15302.482422 -0.070425 -0.071425   \n",
       "4  comp_005959_01W 2003-01-31  39.181721   6355.691406 -0.085813 -0.086813   \n",
       "\n",
       "   ret_exc_lead1m   ret_6_1  ret_12_1  tax_gr1a  ...  cowc_gr1a    pi_nix  \\\n",
       "0       -0.089504 -0.229560 -0.353251 -0.020556  ...   0.041844  3.815513   \n",
       "1        0.108428  0.129149 -0.035109 -0.033306  ...  -0.019596  1.284930   \n",
       "2        0.127227 -0.020286  0.087151 -0.037240  ...   0.026114  1.725004   \n",
       "3        0.064832  0.046064  0.023265 -0.007344  ...  -0.000724  1.961724   \n",
       "4        0.001832  0.123795  0.012811 -0.012241  ...  -0.014214  1.069862   \n",
       "\n",
       "    ret_6_0   ret_1_0    noa_at  cop_atl1  prc_highprc_252d    ocf_at  \\\n",
       "0 -0.244158 -0.018948  0.692706  0.093015          0.398130 -0.007700   \n",
       "1  0.135715  0.005815  0.751569  0.164124          0.918059  0.096164   \n",
       "2 -0.121981 -0.103801  0.961141  0.240563          0.675468  0.070724   \n",
       "3 -0.027605 -0.070425  0.466852  0.221983          0.861333  0.084234   \n",
       "4  0.027359 -0.085813  0.635742  0.108581          0.721256  0.069331   \n",
       "\n",
       "   dbnetis_at  netdebt_me  \n",
       "0    0.085202    0.135368  \n",
       "1    0.063714    0.200036  \n",
       "2   -0.000607    0.410184  \n",
       "3   -0.008142   -0.075232  \n",
       "4   -0.038594    0.500641  \n",
       "\n",
       "[5 rows x 55 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Read in the cleansed and winsorised data\n",
    "df = pd.read_csv(r\"C:\\Users\\krist\\Documents\\Data\\ger_factor_data_from2003.csv\", dtype ={\"comp_tpci\": str}, parse_dates =[\"eom\"])\n",
    "#Convert to float 32 (format needed for the most ML models)\n",
    "df[df.columns[2:]] = df[df.columns[2:]].astype('float32')\n",
    "#Sort observations by date and stock id\n",
    "df = df.sort_values(by = ['eom', 'id'], ascending = True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Copy date variable and stock id to use them as multindex\n",
    "df[\"id2\"] = df[\"id\"].copy()\n",
    "df[\"eom2\"] = df[\"eom\"].copy()\n",
    "df = df.set_index(['eom2','id2'])\n",
    "\n",
    "#Make a copy of  the \"me\" variable (market equity) before rank standartization to use afterwards for value weighting\n",
    "df[\"me2\"] = df[\"me\"].copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To investigate the potential heterogeneity in model predictability, a subgroup analysis for small (the bottom 30% stocks by market equity each month) and large (the top 30% stocks each month) stocks is reported."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Create two new dataframes containing only large and only small firms respectively.\n",
    "p=0.3 \n",
    "df_top = df.groupby('eom').apply(lambda x: x.nlargest(int(len(x)*p),\"me\")).reset_index(drop=True)  \n",
    "df_bottom = df.groupby('eom').apply(lambda x: x.nsmallest(int(len(x)*p),\"me\")).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Rank standardization  in inreval [-1,1]\n",
    "\n",
    "A rank standardization procedure following Kelly, Pruitt, and Su (2019) and Freyberger, Neuhierl, and Weber (2020) is employed. The stock characteristics are ranked month-by-month cross-sectionally and the ranks are mapped into the [-1,1] interval, thereby transforming features into a uniform distribution and increasing the insensitivity to outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Standardize all independent variables\n",
    "features = df.columns[~df.columns.isin([\"eom\",\"eom2\",\"me2\",'id',\"id2\",\"ret\",\"ret_exc\",\"ret_exc_lead1m\"])].tolist()\n",
    "\n",
    "#Ranks in [0,1]interval\n",
    "df[features]=df.groupby(\"eom\")[features].rank(pct=True)\n",
    "\n",
    "#Multiply by 2 and substract 1 to get ranks in interval [-1,1] \n",
    "df[features] = 2*df[features] - 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Standardize the big firms dataset\n",
    "features = df_top.columns[~df_top.columns.isin([\"eom\",\"eom2\",\"me2\",'id',\"id2\",\"ret\",\"ret_exc\",\"ret_exc_lead1m\"])].tolist()\n",
    "df_top[features]=df_top.groupby(\"eom\")[features].rank(pct=True)\n",
    "df_top[features] = 2*df_top[features] - 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "#standardize the small firms dataset\n",
    "features = df_bottom.columns[~df_bottom.columns.isin([\"eom\",\"eom2\",\"me2\",'id',\"id2\",\"ret\",\"ret_exc\",\"ret_exc_lead1m\"])].tolist()\n",
    "df_bottom[features]=df_bottom.groupby(\"eom\")[features].rank(pct=True)\n",
    "df_bottom[features] = 2*df_bottom[features] - 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest\n",
    "\n",
    "### Tuning parameters:\n",
    "\n",
    "\n",
    "- the depth of trees **𝐿**: **max_depth**\n",
    "- the number of trees **𝐵** added to the ensemble prediction: **n_estimators** \n",
    "- the number of predictors **𝑀** randomly considered as potential split variables: **max_features** \n",
    "## All firms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train period: 2003-01-31 - 2008-01-31 ,val period: 2008-01-31 - 2010-01-31 , Test period 2010-01-31 - 2011-01-31 # train records 37757 ,# val records 17356 , # test records 8974\n",
      "Train period: 2004-01-31 - 2009-01-31 ,val period: 2009-01-31 - 2011-01-31 , Test period 2011-01-31 - 2012-01-31 # train records 40925 ,# val records 17143 , # test records 9537\n",
      "Train period: 2005-01-31 - 2010-01-31 ,val period: 2010-01-31 - 2012-01-31 , Test period 2012-01-31 - 2013-01-31 # train records 42278 ,# val records 18511 , # test records 9107\n",
      "Train period: 2006-01-31 - 2011-01-31 ,val period: 2011-01-31 - 2013-01-31 , Test period 2013-01-31 - 2014-01-31 # train records 43971 ,# val records 18644 , # test records 9636\n",
      "Train period: 2007-01-31 - 2012-01-31 ,val period: 2012-01-31 - 2014-01-31 , Test period 2014-01-31 - 2015-01-31 # train records 45267 ,# val records 18743 , # test records 9919\n",
      "Train period: 2008-01-31 - 2013-01-31 ,val period: 2013-01-31 - 2015-01-31 , Test period 2015-01-31 - 2016-01-31 # train records 44974 ,# val records 19555 , # test records 9260\n",
      "Train period: 2009-01-31 - 2014-01-31 ,val period: 2014-01-31 - 2016-01-31 , Test period 2016-01-31 - 2017-01-31 # train records 45423 ,# val records 19179 , # test records 9109\n",
      "Train period: 2010-01-31 - 2015-01-31 ,val period: 2015-01-31 - 2017-01-31 , Test period 2017-01-31 - 2018-01-31 # train records 47173 ,# val records 18369 , # test records 9667\n",
      "Train period: 2011-01-31 - 2016-01-31 ,val period: 2016-01-31 - 2018-01-31 , Test period 2018-01-31 - 2019-01-31 # train records 47459 ,# val records 18776 , # test records 9569\n",
      "Train period: 2012-01-31 - 2017-01-31 ,val period: 2017-01-31 - 2019-01-31 , Test period 2019-01-31 - 2020-01-31 # train records 47031 ,# val records 19236 , # test records 8639\n",
      "R2OOS random forest:  0.008156913790126508\n"
     ]
    }
   ],
   "source": [
    "# Define the splitting scheme:\n",
    "# Rolling window  train-val-test split\n",
    "# The rolling window is set to 96 months (8 years)\n",
    "# 5 years for training, 2 years for validation, and 1 year for out-of-samlpe testing\n",
    "tscv = TimeBasedCV(train_period=60,\n",
    "                   val_period=24,\n",
    "                   test_period=12,\n",
    "                   freq='months')\n",
    "\n",
    "#Define the idependent variables (X) and the dependent variable (y)\n",
    "features = df.columns[~df.columns.isin(['id',\"id2\",\"me2\",\"prc\",\"eom2\",\"ret\",\"ret_exc\",\"ret_exc_lead1m\"])].tolist()\n",
    "X = df[features]\n",
    "y = df[\"ret_exc_lead1m\"]\n",
    "\n",
    "\n",
    "#Empty containers to save results from each window\n",
    "predictions = []\n",
    "y_test_list =[]\n",
    "dates = []\n",
    "dic_r2_all = {}\n",
    "\n",
    "# Model’s complexity: dictionary to save the depth of the trees over time\n",
    "dic_max_depth_all = {}\n",
    "\n",
    "# Grid of prespecified values for each hyperparameter  \n",
    "param_grid = {'max_depth': [1,2,3,4,5,6], 'max_features': [3, 6, 12, 24, 46, 49]}\n",
    "#convert grid to list containing all posible hyperparameter combinations to iterate over\n",
    "grid = list(ParameterGrid(param_grid))\n",
    "# Empty container to save the objective loss function (mean squared error) for each hyperparameter combination\n",
    "mse = np.full((len(grid),1),np.nan, dtype = np.float32)\n",
    "\n",
    "\n",
    "for train_index, val_index, test_index in tscv.split(X, first_split_date= datetime.date(2008,1,31), second_split_date= datetime.date(2010,1,31)):\n",
    "\n",
    "    X_train   = X.loc[train_index].drop('eom', axis=1)\n",
    "    y_train = y.loc[train_index]\n",
    "    \n",
    "    X_val   = X.loc[val_index].drop('eom', axis=1)\n",
    "    y_val = y.loc[val_index]\n",
    "\n",
    "    X_test    = X.loc[test_index].drop('eom', axis=1)\n",
    "    y_test  = y.loc[test_index]\n",
    "    \n",
    "    #Loop over the list containing all pyperparameter combinations, fit on the training sample and use \n",
    "    #validation set to generate predictions:\n",
    "    for i in range(len(grid)):\n",
    "        #n_jobs: number of jobs to run in parallel (-1: use all processors)\n",
    "        # random_state: to ensure that results remain stable\n",
    "        RFR_val = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                                        max_depth= grid[i][\"max_depth\"], \n",
    "                                        max_features=grid[i][\"max_features\"],\n",
    "                                        n_estimators = 100)\n",
    "        RFR_val.fit(X_train, y_train)\n",
    "        Yval_predict=RFR_val.predict(X_val)\n",
    "        #calculate mean squared error for each hyperparameter combination\n",
    "        mse[i,0] = mean_squared_error(y_val,Yval_predict)\n",
    "\n",
    "\n",
    "    #The optimal combination of hyperparameters is the one that causes the lowest loss \n",
    "    optim_param = grid[np.argmin(mse)]\n",
    "\n",
    "    #Fit again using the train and validation set and the optimal value for each hyperparameter \n",
    "    RFR = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                                max_depth=optim_param[\"max_depth\"], \n",
    "                                max_features=optim_param[\"max_features\"],\n",
    "                                n_estimators = 100)\n",
    "    RFR.fit(np.concatenate((X_train, X_val)), np.concatenate((y_train, y_val)))\n",
    "    preds=RFR.predict(X_test)\n",
    "\n",
    "\n",
    "    #Save predictions, dates and the true values of the dependent variable to list\n",
    "    predictions.append(preds)\n",
    "    dates.append(y_test.index)\n",
    "    y_test_list.append(y_test)\n",
    "\n",
    "    #Calculate OOS model performance the for current window\n",
    "    r2 = 1-sum(pow(y_test-preds,2))/sum(pow(y_test,2))\n",
    "    #Save OOS model performance and the respective month to dictionary\n",
    "    dic_r2_all[\"r2.\" + str(y_test.index)] = r2\n",
    "    # Save the depth of the trees to dictionary to explore how model complexity evolves over time\n",
    "    dic_max_depth_all[\"depth.\" + str(y_test.index)]= optim_param[\"max_depth\"]\n",
    "   \n",
    "    \n",
    "    \n",
    "#Concatenate to get results over the whole OOS test period (Jan 2010-Dec 2019)\n",
    "predictions_all= np.concatenate(predictions, axis=0)\n",
    "y_test_list_all= np.concatenate(y_test_list, axis=0) \n",
    "dates_all= np.concatenate(dates, axis=0)\n",
    "\n",
    "R2OOS_RF = 1-sum(pow(y_test_list_all-predictions_all,2))/sum(pow(y_test_list_all,2))\n",
    "print(\"R2OOS random forest: \", R2OOS_RF)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Big only "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train period: 2003-01-31 - 2008-01-31 ,val period: 2008-01-31 - 2010-01-31 , Test period 2010-01-31 - 2011-01-31 # train records 11301 ,# val records 5199 , # test records 2687\n",
      "Train period: 2004-01-31 - 2009-01-31 ,val period: 2009-01-31 - 2011-01-31 , Test period 2011-01-31 - 2012-01-31 # train records 12252 ,# val records 5135 , # test records 2855\n",
      "Train period: 2005-01-31 - 2010-01-31 ,val period: 2010-01-31 - 2012-01-31 , Test period 2012-01-31 - 2013-01-31 # train records 12661 ,# val records 5542 , # test records 2728\n",
      "Train period: 2006-01-31 - 2011-01-31 ,val period: 2011-01-31 - 2013-01-31 , Test period 2013-01-31 - 2014-01-31 # train records 13170 ,# val records 5583 , # test records 2885\n",
      "Train period: 2007-01-31 - 2012-01-31 ,val period: 2012-01-31 - 2014-01-31 , Test period 2014-01-31 - 2015-01-31 # train records 13558 ,# val records 5613 , # test records 2969\n",
      "Train period: 2008-01-31 - 2013-01-31 ,val period: 2013-01-31 - 2015-01-31 , Test period 2015-01-31 - 2016-01-31 # train records 13469 ,# val records 5854 , # test records 2772\n",
      "Train period: 2009-01-31 - 2014-01-31 ,val period: 2014-01-31 - 2016-01-31 , Test period 2016-01-31 - 2017-01-31 # train records 13603 ,# val records 5741 , # test records 2729\n",
      "Train period: 2010-01-31 - 2015-01-31 ,val period: 2015-01-31 - 2017-01-31 , Test period 2017-01-31 - 2018-01-31 # train records 14124 ,# val records 5501 , # test records 2894\n",
      "Train period: 2011-01-31 - 2016-01-31 ,val period: 2016-01-31 - 2018-01-31 , Test period 2018-01-31 - 2019-01-31 # train records 14209 ,# val records 5623 , # test records 2864\n",
      "Train period: 2012-01-31 - 2017-01-31 ,val period: 2017-01-31 - 2019-01-31 , Test period 2019-01-31 - 2020-01-31 # train records 14083 ,# val records 5758 , # test records 2587\n",
      "R2OOS random forest Big only:  0.005369876544611629\n"
     ]
    }
   ],
   "source": [
    "tscv = TimeBasedCV(train_period=60,\n",
    "                   val_period=24,\n",
    "                   test_period=12,\n",
    "                   freq='months')\n",
    "\n",
    "features = df_top.columns[~df_top.columns.isin(['id',\"id2\",\"me2\",\"prc\",\"eom2\",\"ret\",\"ret_exc\",\"ret_exc_lead1m\"])].tolist()\n",
    "X = df_top[features]\n",
    "y = df_top[\"ret_exc_lead1m\"]\n",
    "\n",
    "\n",
    "\n",
    "predictions_top = []\n",
    "y_test_list_top =[]\n",
    "dates_top = []\n",
    "dic_r2_all_top = {}\n",
    "\n",
    "\n",
    "param_grid = {'max_depth': [1,2,3,4,5,6], 'max_features': [3, 6, 12, 24, 46, 49]}\n",
    "grid = list(ParameterGrid(param_grid))\n",
    "mse = np.full((len(grid),1),np.nan, dtype = np.float32)\n",
    "\n",
    "\n",
    "for train_index, val_index, test_index in tscv.split(X, first_split_date= datetime.date(2008,1,31), second_split_date= datetime.date(2010,1,31)):\n",
    "\n",
    "    X_train   = X.loc[train_index].drop('eom', axis=1)\n",
    "    y_train = y.loc[train_index]\n",
    "    \n",
    "    X_val   = X.loc[val_index].drop('eom', axis=1)\n",
    "    y_val = y.loc[val_index]\n",
    "\n",
    "    X_test    = X.loc[test_index].drop('eom', axis=1)\n",
    "    y_test  = y.loc[test_index]\n",
    "    \n",
    "    for i in range(len(grid)):\n",
    "        RFR_val = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                                        max_depth= grid[i][\"max_depth\"], \n",
    "                                        max_features=grid[i][\"max_features\"],\n",
    "                                        n_estimators = 100)\n",
    "        RFR_val.fit(X_train, y_train)\n",
    "        Yval_predict=RFR_val.predict(X_val)\n",
    "        mse[i,0] = mean_squared_error(y_val,Yval_predict)\n",
    "\n",
    "    optim_param = grid[np.argmin(mse)]\n",
    "\n",
    "    RFR = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                                max_depth=optim_param[\"max_depth\"], \n",
    "                                max_features=optim_param[\"max_features\"],\n",
    "                                n_estimators = 100)\n",
    "    RFR.fit(np.concatenate((X_train, X_val)), np.concatenate((y_train, y_val)))\n",
    "    preds=RFR.predict(X_test)\n",
    "\n",
    "\n",
    "\n",
    "    predictions_top.append(preds)\n",
    "    dates_top.append(y_test.index)\n",
    "    y_test_list_top.append(y_test)\n",
    "\n",
    "    \n",
    "    \n",
    "predictions_all_top= np.concatenate(predictions_top, axis=0)\n",
    "y_test_list_all_top= np.concatenate(y_test_list_top, axis=0) \n",
    "dates_all_top= np.concatenate(dates_top, axis=0)\n",
    "\n",
    "R2OOS_RF_top = 1-sum(pow(y_test_list_all_top-predictions_all_top,2))/sum(pow(y_test_list_all_top,2))\n",
    "print(\"R2OOS random forest Big only: \", R2OOS_RF_top)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Small only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train period: 2003-01-31 - 2008-01-31 ,val period: 2008-01-31 - 2010-01-31 , Test period 2010-01-31 - 2011-01-31 # train records 11301 ,# val records 5199 , # test records 2687\n",
      "Train period: 2004-01-31 - 2009-01-31 ,val period: 2009-01-31 - 2011-01-31 , Test period 2011-01-31 - 2012-01-31 # train records 12252 ,# val records 5135 , # test records 2855\n",
      "Train period: 2005-01-31 - 2010-01-31 ,val period: 2010-01-31 - 2012-01-31 , Test period 2012-01-31 - 2013-01-31 # train records 12661 ,# val records 5542 , # test records 2728\n",
      "Train period: 2006-01-31 - 2011-01-31 ,val period: 2011-01-31 - 2013-01-31 , Test period 2013-01-31 - 2014-01-31 # train records 13170 ,# val records 5583 , # test records 2885\n",
      "Train period: 2007-01-31 - 2012-01-31 ,val period: 2012-01-31 - 2014-01-31 , Test period 2014-01-31 - 2015-01-31 # train records 13558 ,# val records 5613 , # test records 2969\n",
      "Train period: 2008-01-31 - 2013-01-31 ,val period: 2013-01-31 - 2015-01-31 , Test period 2015-01-31 - 2016-01-31 # train records 13469 ,# val records 5854 , # test records 2772\n",
      "Train period: 2009-01-31 - 2014-01-31 ,val period: 2014-01-31 - 2016-01-31 , Test period 2016-01-31 - 2017-01-31 # train records 13603 ,# val records 5741 , # test records 2729\n",
      "Train period: 2010-01-31 - 2015-01-31 ,val period: 2015-01-31 - 2017-01-31 , Test period 2017-01-31 - 2018-01-31 # train records 14124 ,# val records 5501 , # test records 2894\n",
      "Train period: 2011-01-31 - 2016-01-31 ,val period: 2016-01-31 - 2018-01-31 , Test period 2018-01-31 - 2019-01-31 # train records 14209 ,# val records 5623 , # test records 2864\n",
      "Train period: 2012-01-31 - 2017-01-31 ,val period: 2017-01-31 - 2019-01-31 , Test period 2019-01-31 - 2020-01-31 # train records 14083 ,# val records 5758 , # test records 2587\n",
      "R2OOS random forest Big only:  0.008501380189670393\n"
     ]
    }
   ],
   "source": [
    "tscv = TimeBasedCV(train_period=60,\n",
    "                   val_period=24,\n",
    "                   test_period=12,\n",
    "                   freq='months')\n",
    "\n",
    "features = df_bottom.columns[~df_bottom.columns.isin(['id',\"id2\",\"me2\",\"prc\",\"eom2\",\"ret\",\"ret_exc\",\"ret_exc_lead1m\"])].tolist()\n",
    "X = df_bottom[features]\n",
    "y = df_bottom[\"ret_exc_lead1m\"]\n",
    "\n",
    "\n",
    "\n",
    "predictions_bottom = []\n",
    "y_test_list_bottom =[]\n",
    "dates_bottom = []\n",
    "dic_r2_all_bottom = {}\n",
    "\n",
    "\n",
    "param_grid = {'max_depth': [1,2,3,4,5,6], 'max_features': [3, 6, 12, 24, 46, 49]}\n",
    "grid = list(ParameterGrid(param_grid))\n",
    "mse = np.full((len(grid),1),np.nan, dtype = np.float32)\n",
    "\n",
    "\n",
    "for train_index, val_index, test_index in tscv.split(X, first_split_date= datetime.date(2008,1,31), second_split_date= datetime.date(2010,1,31)):\n",
    "\n",
    "    X_train   = X.loc[train_index].drop('eom', axis=1)\n",
    "    y_train = y.loc[train_index]\n",
    "    \n",
    "    X_val   = X.loc[val_index].drop('eom', axis=1)\n",
    "    y_val = y.loc[val_index]\n",
    "\n",
    "    X_test    = X.loc[test_index].drop('eom', axis=1)\n",
    "    y_test  = y.loc[test_index]\n",
    "    \n",
    "    for i in range(len(grid)):\n",
    "        RFR_val = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                                        max_depth= grid[i][\"max_depth\"], \n",
    "                                        max_features=grid[i][\"max_features\"],\n",
    "                                        n_estimators = 100)\n",
    "        RFR_val.fit(X_train, y_train)\n",
    "        Yval_predict=RFR_val.predict(X_val)\n",
    "        mse[i,0] = mean_squared_error(y_val,Yval_predict)\n",
    "\n",
    "    optim_param = grid[np.argmin(mse)]\n",
    "\n",
    "    RFR = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                                max_depth=optim_param[\"max_depth\"], \n",
    "                                max_features=optim_param[\"max_features\"],\n",
    "                                n_estimators = 100)\n",
    "    RFR.fit(np.concatenate((X_train, X_val)), np.concatenate((y_train, y_val)))\n",
    "    preds=RFR.predict(X_test)\n",
    "\n",
    "\n",
    "\n",
    "    predictions_bottom.append(preds)\n",
    "    dates_bottom.append(y_test.index)\n",
    "    y_test_list_bottom.append(y_test)\n",
    "\n",
    "    \n",
    "    \n",
    "predictions_all_bottom= np.concatenate(predictions_bottom, axis=0)\n",
    "y_test_list_all_bottom= np.concatenate(y_test_list_bottom, axis=0) \n",
    "dates_all_bottom= np.concatenate(dates_bottom, axis=0)\n",
    "\n",
    "R2OOS_RF_bottom = 1-sum(pow(y_test_list_all_bottom-predictions_all_bottom,2))/sum(pow(y_test_list_all_bottom,2))\n",
    "print(\"R2OOS random forest Big only: \", R2OOS_RF_bottom)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>RF</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Full sample</th>\n",
       "      <td>0.008157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Large firms</th>\n",
       "      <td>0.005370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Small firms</th>\n",
       "      <td>0.008501</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   RF\n",
       "Full sample  0.008157\n",
       "Large firms  0.005370\n",
       "Small firms  0.008501"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Generate table containing the R2oos measures for all three datasets\n",
    "chart = np.array([[R2OOS_RF],\n",
    "                  [R2OOS_RF_top],\n",
    "                  [R2OOS_RF_bottom]])\n",
    "                     \n",
    "\n",
    "r2oos_rf = pd.DataFrame(chart, columns=['RF'],\n",
    "                              index=[\"Full sample\", \"Large firms\", \"Small firms\"])\n",
    "r2oos_rf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Save the model's performance measures to compare with other models later.\n",
    "r2oos_rf.to_csv(r'C:\\Users\\krist\\Documents\\Data\\r2oos results\\r2oos_rf.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Time-varying model complexity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Convert dictionary containing the number of components over time in dataframe\n",
    "pd.DataFrame(dic_max_depth_all.items())\n",
    "depth =pd.DataFrame(dic_max_depth_all.items(), columns=['Identifier', 'max_depth'])\n",
    "depth['Identifier'] = depth['Identifier'].astype(str)\n",
    "depth[\"time\"] = depth[\"Identifier\"].str[20:30]\n",
    "depth.drop([\"Identifier\"], axis = 1, inplace = True)\n",
    "depth[\"time\"] = pd.to_datetime(depth[\"time\"], utc = False)\n",
    "# Plot time-varying model complexity\n",
    "depth.set_index('time').plot();\n",
    "plt.xticks(rotation=45);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Save the model's complexity  to compare with other models later.\n",
    "depth.to_csv(r'C:\\Users\\krist\\Documents\\Data\\model_complexity\\comp_rf.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Variable Importance\n",
    "\n",
    "Variable importance is defined following Gu et al. (2020), i.e., for each model, all values of a given predictor are set to zero and the reduction in predictive R2OOS is calculated. Then, the absolute reductions in predictive R2OOS are normalized to sum to 1, indicating the relative contribution of each variable to a model. In contrast to Gu et al. (2020), variable importance is calculated based on the last rolling window observation sample and does not represent an average over the whole sample.  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.021630052470213545\n"
     ]
    }
   ],
   "source": [
    "# R2OOS based on the last rolling window when all variables are included\n",
    "features = df.columns[~df.columns.isin(['id',\"id2\",\"eom\",\"eom2\",\"me2\",\"prc\",\"ret\",\"ret_exc\",\"ret_exc_lead1m\",\"year\"])].tolist()\n",
    "df[\"year\"] = df[\"eom\"].dt.year\n",
    "\n",
    "X_train = df[features].loc[(df[\"year\"]>=2012) & (df[\"year\"]<=2016)]\n",
    "y_train = df[\"ret_exc_lead1m\"].loc[(df[\"year\"]>=2012) & (df[\"year\"]<=2016)]\n",
    "\n",
    "X_val = df[features].loc[(df[\"year\"]>=2017) & (df[\"year\"]<=2018)]\n",
    "y_val = df[\"ret_exc_lead1m\"].loc[(df[\"year\"]>=2017) & (df[\"year\"]<=2018)]\n",
    "\n",
    "param_grid = {'max_depth': [1,2,3,4,5,6], 'max_features': [3, 6, 12, 24, 46, 49]}\n",
    "grid = list(ParameterGrid(param_grid))\n",
    "mse = np.full((len(grid),1),np.nan, dtype = np.float32)\n",
    "\n",
    "for i in range(len(grid)):\n",
    "    RFR_val = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                                    max_depth= grid[i][\"max_depth\"], \n",
    "                                    max_features=grid[i][\"max_features\"],\n",
    "                                    n_estimators = 100)\n",
    "    RFR_val.fit(X_train, y_train)\n",
    "    Yval_predict=RFR_val.predict(X_val)\n",
    "    mse[i,0] = mean_squared_error(y_val,Yval_predict)\n",
    "\n",
    "optim_param = grid[np.argmin(mse)]\n",
    "RFR = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                            max_depth=optim_param[\"max_depth\"], \n",
    "                            max_features=optim_param[\"max_features\"],\n",
    "                            n_estimators = 100)\n",
    "RFR.fit(np.concatenate((X_train, X_val)), np.concatenate((y_train, y_val)))\n",
    "preds=RFR.predict(np.concatenate((X_train, X_val)))\n",
    "R2OOS_all = 1-sum(pow(np.concatenate((y_train, y_val))-preds,2))/sum(pow(np.concatenate((y_train, y_val)),2))\n",
    "print(R2OOS_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate a separate dataframe for each independent variable, where all values of that variable are set to zero\n",
    "\n",
    "for j in features:\n",
    "    globals()['df_' + str(j)] =  df.copy()\n",
    "    globals()['df_' + str(j)][str(j)] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>eom</th>\n",
       "      <th>prc</th>\n",
       "      <th>me</th>\n",
       "      <th>ret</th>\n",
       "      <th>ret_exc</th>\n",
       "      <th>ret_exc_lead1m</th>\n",
       "      <th>ret_6_1</th>\n",
       "      <th>ret_12_1</th>\n",
       "      <th>tax_gr1a</th>\n",
       "      <th>...</th>\n",
       "      <th>ret_6_0</th>\n",
       "      <th>ret_1_0</th>\n",
       "      <th>noa_at</th>\n",
       "      <th>cop_atl1</th>\n",
       "      <th>prc_highprc_252d</th>\n",
       "      <th>ocf_at</th>\n",
       "      <th>dbnetis_at</th>\n",
       "      <th>netdebt_me</th>\n",
       "      <th>me2</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>eom2</th>\n",
       "      <th>id2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2003-01-31</th>\n",
       "      <th>comp_001166_02W</th>\n",
       "      <td>comp_001166_02W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>-0.376652</td>\n",
       "      <td>0.162996</td>\n",
       "      <td>-0.018948</td>\n",
       "      <td>-0.019948</td>\n",
       "      <td>-0.089504</td>\n",
       "      <td>-0.665198</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.726872</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.669604</td>\n",
       "      <td>-0.312775</td>\n",
       "      <td>0.585903</td>\n",
       "      <td>-0.735683</td>\n",
       "      <td>-0.859031</td>\n",
       "      <td>-0.797357</td>\n",
       "      <td>0.735683</td>\n",
       "      <td>-0.149780</td>\n",
       "      <td>567.074646</td>\n",
       "      <td>2003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_001661_01W</th>\n",
       "      <td>comp_001661_01W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>0.427313</td>\n",
       "      <td>0.718062</td>\n",
       "      <td>0.005815</td>\n",
       "      <td>0.004815</td>\n",
       "      <td>0.108428</td>\n",
       "      <td>0.669604</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.845815</td>\n",
       "      <td>...</td>\n",
       "      <td>0.577093</td>\n",
       "      <td>-0.110132</td>\n",
       "      <td>0.696035</td>\n",
       "      <td>-0.392070</td>\n",
       "      <td>0.607930</td>\n",
       "      <td>0.286344</td>\n",
       "      <td>0.647577</td>\n",
       "      <td>0.145374</td>\n",
       "      <td>5221.931641</td>\n",
       "      <td>2003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_004367_02W</th>\n",
       "      <td>comp_004367_02W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>0.433921</td>\n",
       "      <td>0.674009</td>\n",
       "      <td>-0.103801</td>\n",
       "      <td>-0.104801</td>\n",
       "      <td>0.127227</td>\n",
       "      <td>0.114537</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.894273</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.356828</td>\n",
       "      <td>-0.766520</td>\n",
       "      <td>0.881057</td>\n",
       "      <td>0.458150</td>\n",
       "      <td>-0.365639</td>\n",
       "      <td>-0.110132</td>\n",
       "      <td>-0.207048</td>\n",
       "      <td>0.339207</td>\n",
       "      <td>4139.599121</td>\n",
       "      <td>2003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_004925_02W</th>\n",
       "      <td>comp_004925_02W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>0.295154</td>\n",
       "      <td>0.881057</td>\n",
       "      <td>-0.070425</td>\n",
       "      <td>-0.071425</td>\n",
       "      <td>0.064832</td>\n",
       "      <td>0.392070</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.453744</td>\n",
       "      <td>...</td>\n",
       "      <td>0.096916</td>\n",
       "      <td>-0.612335</td>\n",
       "      <td>-0.264317</td>\n",
       "      <td>0.330396</td>\n",
       "      <td>0.422907</td>\n",
       "      <td>0.149780</td>\n",
       "      <td>-0.370044</td>\n",
       "      <td>-0.709251</td>\n",
       "      <td>15302.482422</td>\n",
       "      <td>2003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_005959_01W</th>\n",
       "      <td>comp_005959_01W</td>\n",
       "      <td>2003-01-31</td>\n",
       "      <td>0.477974</td>\n",
       "      <td>0.762115</td>\n",
       "      <td>-0.085813</td>\n",
       "      <td>-0.086813</td>\n",
       "      <td>0.001832</td>\n",
       "      <td>0.643172</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.599119</td>\n",
       "      <td>...</td>\n",
       "      <td>0.259912</td>\n",
       "      <td>-0.691630</td>\n",
       "      <td>0.458150</td>\n",
       "      <td>-0.669604</td>\n",
       "      <td>-0.255507</td>\n",
       "      <td>-0.118943</td>\n",
       "      <td>-0.656388</td>\n",
       "      <td>0.436123</td>\n",
       "      <td>6355.691406</td>\n",
       "      <td>2003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2020-12-31</th>\n",
       "      <th>comp_333885_01W</th>\n",
       "      <td>comp_333885_01W</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>0.367156</td>\n",
       "      <td>0.484646</td>\n",
       "      <td>0.128645</td>\n",
       "      <td>0.128545</td>\n",
       "      <td>-0.037398</td>\n",
       "      <td>-0.957276</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.951936</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.853138</td>\n",
       "      <td>0.591455</td>\n",
       "      <td>0.457944</td>\n",
       "      <td>0.815754</td>\n",
       "      <td>-0.319092</td>\n",
       "      <td>0.877170</td>\n",
       "      <td>-0.930574</td>\n",
       "      <td>-0.316422</td>\n",
       "      <td>10829.783203</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_334302_02W</th>\n",
       "      <td>comp_334302_02W</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>-0.842457</td>\n",
       "      <td>-0.487316</td>\n",
       "      <td>0.097961</td>\n",
       "      <td>0.097861</td>\n",
       "      <td>-0.033670</td>\n",
       "      <td>-0.420561</td>\n",
       "      <td>0</td>\n",
       "      <td>0.540721</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.268358</td>\n",
       "      <td>0.417891</td>\n",
       "      <td>0.791722</td>\n",
       "      <td>0.345794</td>\n",
       "      <td>0.754339</td>\n",
       "      <td>0.054740</td>\n",
       "      <td>0.692924</td>\n",
       "      <td>-0.252336</td>\n",
       "      <td>297.482697</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_335706_02W</th>\n",
       "      <td>comp_335706_02W</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>-0.722296</td>\n",
       "      <td>-0.618158</td>\n",
       "      <td>0.114381</td>\n",
       "      <td>0.114281</td>\n",
       "      <td>0.089684</td>\n",
       "      <td>0.097463</td>\n",
       "      <td>0</td>\n",
       "      <td>0.311081</td>\n",
       "      <td>...</td>\n",
       "      <td>0.236315</td>\n",
       "      <td>0.511348</td>\n",
       "      <td>-0.241656</td>\n",
       "      <td>-0.935915</td>\n",
       "      <td>-0.313752</td>\n",
       "      <td>-0.917223</td>\n",
       "      <td>0.164219</td>\n",
       "      <td>-0.730307</td>\n",
       "      <td>196.243454</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_340115_01W</th>\n",
       "      <td>comp_340115_01W</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>-0.392523</td>\n",
       "      <td>-0.025367</td>\n",
       "      <td>0.155813</td>\n",
       "      <td>0.155713</td>\n",
       "      <td>0.009796</td>\n",
       "      <td>0.001335</td>\n",
       "      <td>0</td>\n",
       "      <td>0.871829</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001335</td>\n",
       "      <td>0.687583</td>\n",
       "      <td>-0.415220</td>\n",
       "      <td>-0.436582</td>\n",
       "      <td>0.001335</td>\n",
       "      <td>0.025367</td>\n",
       "      <td>0.642190</td>\n",
       "      <td>0.562083</td>\n",
       "      <td>1799.958496</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_340153_01W</th>\n",
       "      <td>comp_340153_01W</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>0.138852</td>\n",
       "      <td>0.711615</td>\n",
       "      <td>0.238817</td>\n",
       "      <td>0.238717</td>\n",
       "      <td>0.006943</td>\n",
       "      <td>0.001335</td>\n",
       "      <td>0</td>\n",
       "      <td>0.847797</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001335</td>\n",
       "      <td>0.879840</td>\n",
       "      <td>-0.815754</td>\n",
       "      <td>-0.636849</td>\n",
       "      <td>0.001335</td>\n",
       "      <td>-0.129506</td>\n",
       "      <td>-0.110814</td>\n",
       "      <td>-0.578104</td>\n",
       "      <td>26788.144531</td>\n",
       "      <td>2020</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>156948 rows × 57 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         id        eom       prc        me  \\\n",
       "eom2       id2                                                               \n",
       "2003-01-31 comp_001166_02W  comp_001166_02W 2003-01-31 -0.376652  0.162996   \n",
       "           comp_001661_01W  comp_001661_01W 2003-01-31  0.427313  0.718062   \n",
       "           comp_004367_02W  comp_004367_02W 2003-01-31  0.433921  0.674009   \n",
       "           comp_004925_02W  comp_004925_02W 2003-01-31  0.295154  0.881057   \n",
       "           comp_005959_01W  comp_005959_01W 2003-01-31  0.477974  0.762115   \n",
       "...                                     ...        ...       ...       ...   \n",
       "2020-12-31 comp_333885_01W  comp_333885_01W 2020-12-31  0.367156  0.484646   \n",
       "           comp_334302_02W  comp_334302_02W 2020-12-31 -0.842457 -0.487316   \n",
       "           comp_335706_02W  comp_335706_02W 2020-12-31 -0.722296 -0.618158   \n",
       "           comp_340115_01W  comp_340115_01W 2020-12-31 -0.392523 -0.025367   \n",
       "           comp_340153_01W  comp_340153_01W 2020-12-31  0.138852  0.711615   \n",
       "\n",
       "                                 ret   ret_exc  ret_exc_lead1m   ret_6_1  \\\n",
       "eom2       id2                                                             \n",
       "2003-01-31 comp_001166_02W -0.018948 -0.019948       -0.089504 -0.665198   \n",
       "           comp_001661_01W  0.005815  0.004815        0.108428  0.669604   \n",
       "           comp_004367_02W -0.103801 -0.104801        0.127227  0.114537   \n",
       "           comp_004925_02W -0.070425 -0.071425        0.064832  0.392070   \n",
       "           comp_005959_01W -0.085813 -0.086813        0.001832  0.643172   \n",
       "...                              ...       ...             ...       ...   \n",
       "2020-12-31 comp_333885_01W  0.128645  0.128545       -0.037398 -0.957276   \n",
       "           comp_334302_02W  0.097961  0.097861       -0.033670 -0.420561   \n",
       "           comp_335706_02W  0.114381  0.114281        0.089684  0.097463   \n",
       "           comp_340115_01W  0.155813  0.155713        0.009796  0.001335   \n",
       "           comp_340153_01W  0.238817  0.238717        0.006943  0.001335   \n",
       "\n",
       "                            ret_12_1  tax_gr1a  ...   ret_6_0   ret_1_0  \\\n",
       "eom2       id2                                  ...                       \n",
       "2003-01-31 comp_001166_02W         0 -0.726872  ... -0.669604 -0.312775   \n",
       "           comp_001661_01W         0 -0.845815  ...  0.577093 -0.110132   \n",
       "           comp_004367_02W         0 -0.894273  ... -0.356828 -0.766520   \n",
       "           comp_004925_02W         0 -0.453744  ...  0.096916 -0.612335   \n",
       "           comp_005959_01W         0 -0.599119  ...  0.259912 -0.691630   \n",
       "...                              ...       ...  ...       ...       ...   \n",
       "2020-12-31 comp_333885_01W         0 -0.951936  ... -0.853138  0.591455   \n",
       "           comp_334302_02W         0  0.540721  ... -0.268358  0.417891   \n",
       "           comp_335706_02W         0  0.311081  ...  0.236315  0.511348   \n",
       "           comp_340115_01W         0  0.871829  ...  0.001335  0.687583   \n",
       "           comp_340153_01W         0  0.847797  ...  0.001335  0.879840   \n",
       "\n",
       "                              noa_at  cop_atl1  prc_highprc_252d    ocf_at  \\\n",
       "eom2       id2                                                               \n",
       "2003-01-31 comp_001166_02W  0.585903 -0.735683         -0.859031 -0.797357   \n",
       "           comp_001661_01W  0.696035 -0.392070          0.607930  0.286344   \n",
       "           comp_004367_02W  0.881057  0.458150         -0.365639 -0.110132   \n",
       "           comp_004925_02W -0.264317  0.330396          0.422907  0.149780   \n",
       "           comp_005959_01W  0.458150 -0.669604         -0.255507 -0.118943   \n",
       "...                              ...       ...               ...       ...   \n",
       "2020-12-31 comp_333885_01W  0.457944  0.815754         -0.319092  0.877170   \n",
       "           comp_334302_02W  0.791722  0.345794          0.754339  0.054740   \n",
       "           comp_335706_02W -0.241656 -0.935915         -0.313752 -0.917223   \n",
       "           comp_340115_01W -0.415220 -0.436582          0.001335  0.025367   \n",
       "           comp_340153_01W -0.815754 -0.636849          0.001335 -0.129506   \n",
       "\n",
       "                            dbnetis_at  netdebt_me           me2  year  \n",
       "eom2       id2                                                          \n",
       "2003-01-31 comp_001166_02W    0.735683   -0.149780    567.074646  2003  \n",
       "           comp_001661_01W    0.647577    0.145374   5221.931641  2003  \n",
       "           comp_004367_02W   -0.207048    0.339207   4139.599121  2003  \n",
       "           comp_004925_02W   -0.370044   -0.709251  15302.482422  2003  \n",
       "           comp_005959_01W   -0.656388    0.436123   6355.691406  2003  \n",
       "...                                ...         ...           ...   ...  \n",
       "2020-12-31 comp_333885_01W   -0.930574   -0.316422  10829.783203  2020  \n",
       "           comp_334302_02W    0.692924   -0.252336    297.482697  2020  \n",
       "           comp_335706_02W    0.164219   -0.730307    196.243454  2020  \n",
       "           comp_340115_01W    0.642190    0.562083   1799.958496  2020  \n",
       "           comp_340153_01W   -0.110814   -0.578104  26788.144531  2020  \n",
       "\n",
       "[156948 rows x 57 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#e.g., df_ret_12_1 is a dataframe where all values of the ret_12_1 variable are set to zero\n",
    "df_ret_12_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate R2OOS based on the last rolling window for each dataframe and save to dictionary\n",
    "\n",
    "dic = {}    \n",
    "param_grid = {'max_depth': [1,2,3,4,5,6], 'max_features': [3, 6, 12, 24, 46, 49]}\n",
    "grid = list(ParameterGrid(param_grid))\n",
    "mse = np.full((len(grid),1),np.nan, dtype = np.float32)\n",
    "\n",
    "    \n",
    "for j in features:\n",
    "    df_var = globals()['df_' + str(j)]\n",
    "    \n",
    "    X_train = df_var[features].loc[(df_var[\"year\"]>=2012) & (df_var[\"year\"]<=2016)]\n",
    "    y_train = df_var[\"ret_exc_lead1m\"].loc[(df_var[\"year\"]>=2012) & (df_var[\"year\"]<=2016)]\n",
    "\n",
    "    X_val = df_var[features].loc[(df_var[\"year\"]>=2017) & (df_var[\"year\"]<=2018)]\n",
    "    y_val = df_var[\"ret_exc_lead1m\"].loc[(df_var[\"year\"]>=2017) & (df_var[\"year\"]<=2018)]\n",
    "\n",
    "    X_test = df_var[features].loc[(df_var[\"year\"]>=2019) & (df_var[\"year\"]<2020)]\n",
    "    y_test = df_var[\"ret_exc_lead1m\"].loc[(df_var[\"year\"]>=2019) & (df_var[\"year\"]<2020)]\n",
    "\n",
    " \n",
    "    for i in range(len(grid)):\n",
    "        RFR_val = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                                        max_depth= grid[i][\"max_depth\"], \n",
    "                                        max_features=grid[i][\"max_features\"],\n",
    "                                        n_estimators = 100)\n",
    "        RFR_val.fit(X_train, y_train)\n",
    "        Yval_predict=RFR_val.predict(X_val)\n",
    "        mse[i,0] = mean_squared_error(y_val,Yval_predict)\n",
    "\n",
    "    optim_param = grid[np.argmin(mse)]\n",
    "    RFR = RandomForestRegressor(bootstrap = True,random_state=42, n_jobs=-1,\n",
    "                                max_depth=optim_param[\"max_depth\"], \n",
    "                                max_features=optim_param[\"max_features\"],\n",
    "                                n_estimators = 100)\n",
    "    RFR.fit(np.concatenate((X_train, X_val)), np.concatenate((y_train, y_val)))\n",
    "    preds=RFR.predict(np.concatenate((X_train, X_val)))\n",
    "    R2OOS_var = 1-sum(pow(np.concatenate((y_train, y_val))-preds,2))/sum(pow(np.concatenate((y_train, y_val)),2))\n",
    "    dic['R2OOS_' + str(j)] = R2OOS_var"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'R2OOS_me': 0.021413656433947992,\n",
       " 'R2OOS_ret_6_1': 0.0220330921659746,\n",
       " 'R2OOS_ret_12_1': 0.02063175822966734,\n",
       " 'R2OOS_tax_gr1a': 0.020800719372490062,\n",
       " 'R2OOS_be_me': 0.02179083108827007,\n",
       " 'R2OOS_debt_me': 0.02155544328886516,\n",
       " 'R2OOS_ni_me': 0.021649235358731866,\n",
       " 'R2OOS_sale_gr3': 0.020980667433348366,\n",
       " 'R2OOS_sale_gr1': 0.021000325790497265,\n",
       " 'R2OOS_sale_me': 0.028714122659467933,\n",
       " 'R2OOS_lnoa_gr1a': 0.021331350397952642,\n",
       " 'R2OOS_inv_gr1a': 0.021157716227448886,\n",
       " 'R2OOS_oaccruals_at': 0.02084613607145347,\n",
       " 'R2OOS_taccruals_at': 0.020701710882404134,\n",
       " 'R2OOS_be_gr1a': 0.02140182010987335,\n",
       " 'R2OOS_taccruals_ni': 0.02173738886849652,\n",
       " 'R2OOS_ebit_sale': 0.02201346292725448,\n",
       " 'R2OOS_sale_bev': 0.028501047684055814,\n",
       " 'R2OOS_age': 0.015603821622703462,\n",
       " 'R2OOS_beta_60m': 0.029103876372342796,\n",
       " 'R2OOS_rmax1_21d': 0.021883380247232354,\n",
       " 'R2OOS_bidaskhl_21d': 0.022154065362073205,\n",
       " 'R2OOS_ret_9_1': 0.027864518640730807,\n",
       " 'R2OOS_ret_12_7': 0.021515045586456116,\n",
       " 'R2OOS_ni_be': 0.021745577093492008,\n",
       " 'R2OOS_ocf_debt': 0.022012639164640024,\n",
       " 'R2OOS_cash_gr1a': 0.021529336333269034,\n",
       " 'R2OOS_at_gr1': 0.028664958741197233,\n",
       " 'R2OOS_chcsho_12m': 0.019390526081093706,\n",
       " 'R2OOS_ca_cl': 0.02197658368508526,\n",
       " 'R2OOS_ivol_capm_252d': 0.02160857425484075,\n",
       " 'R2OOS_debt_gr1': 0.028386890842465706,\n",
       " 'R2OOS_debtlt_gr1a': 0.028606836088023946,\n",
       " 'R2OOS_ope_be': 0.02118707047027324,\n",
       " 'R2OOS_caliq_cl': 0.02198673576338106,\n",
       " 'R2OOS_dsale_drec': 0.028190363424244747,\n",
       " 'R2OOS_f_score': 0.02785147224767559,\n",
       " 'R2OOS_rvol_21d': 0.021226221467178008,\n",
       " 'R2OOS_gp_bev': 0.02068042761491873,\n",
       " 'R2OOS_cowc_gr1a': 0.02167182477166918,\n",
       " 'R2OOS_pi_nix': 0.022024767207282747,\n",
       " 'R2OOS_ret_6_0': 0.021157578476293626,\n",
       " 'R2OOS_ret_1_0': 0.021793276203512124,\n",
       " 'R2OOS_noa_at': 0.022133076052234357,\n",
       " 'R2OOS_cop_atl1': 0.028252188935828926,\n",
       " 'R2OOS_prc_highprc_252d': 0.03677927306891149,\n",
       " 'R2OOS_ocf_at': 0.021079277150913844,\n",
       " 'R2OOS_dbnetis_at': 0.02196286620502208,\n",
       " 'R2OOS_netdebt_me': 0.02938252492325588}"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dictionary, containing OOS predictive R2, when the all values of a given predictor are set to zero\n",
    "dic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Feature</th>\n",
       "      <th>R2OOS</th>\n",
       "      <th>red_R2OOS</th>\n",
       "      <th>var_imp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>prc_highprc_252d</td>\n",
       "      <td>0.036779</td>\n",
       "      <td>-0.015149</td>\n",
       "      <td>0.194736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>netdebt_me</td>\n",
       "      <td>0.029383</td>\n",
       "      <td>-0.007752</td>\n",
       "      <td>0.099655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>beta_60m</td>\n",
       "      <td>0.029104</td>\n",
       "      <td>-0.007474</td>\n",
       "      <td>0.096073</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>sale_me</td>\n",
       "      <td>0.028714</td>\n",
       "      <td>-0.007084</td>\n",
       "      <td>0.091063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>at_gr1</td>\n",
       "      <td>0.028665</td>\n",
       "      <td>-0.007035</td>\n",
       "      <td>0.090431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>debtlt_gr1a</td>\n",
       "      <td>0.028607</td>\n",
       "      <td>-0.006977</td>\n",
       "      <td>0.089683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>sale_bev</td>\n",
       "      <td>0.028501</td>\n",
       "      <td>-0.006871</td>\n",
       "      <td>0.088324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>debt_gr1</td>\n",
       "      <td>0.028387</td>\n",
       "      <td>-0.006757</td>\n",
       "      <td>0.086856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>cop_atl1</td>\n",
       "      <td>0.028252</td>\n",
       "      <td>-0.006622</td>\n",
       "      <td>0.085125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>dsale_drec</td>\n",
       "      <td>0.028190</td>\n",
       "      <td>-0.006560</td>\n",
       "      <td>0.084330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>ret_9_1</td>\n",
       "      <td>0.027865</td>\n",
       "      <td>-0.006234</td>\n",
       "      <td>0.080141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>f_score</td>\n",
       "      <td>0.027851</td>\n",
       "      <td>-0.006221</td>\n",
       "      <td>0.079974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>bidaskhl_21d</td>\n",
       "      <td>0.022154</td>\n",
       "      <td>-0.000524</td>\n",
       "      <td>0.006736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>noa_at</td>\n",
       "      <td>0.022133</td>\n",
       "      <td>-0.000503</td>\n",
       "      <td>0.006466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ret_6_1</td>\n",
       "      <td>0.022033</td>\n",
       "      <td>-0.000403</td>\n",
       "      <td>0.005181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>pi_nix</td>\n",
       "      <td>0.022025</td>\n",
       "      <td>-0.000395</td>\n",
       "      <td>0.005074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>ebit_sale</td>\n",
       "      <td>0.022013</td>\n",
       "      <td>-0.000383</td>\n",
       "      <td>0.004929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>ocf_debt</td>\n",
       "      <td>0.022013</td>\n",
       "      <td>-0.000383</td>\n",
       "      <td>0.004918</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>caliq_cl</td>\n",
       "      <td>0.021987</td>\n",
       "      <td>-0.000357</td>\n",
       "      <td>0.004585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>ca_cl</td>\n",
       "      <td>0.021977</td>\n",
       "      <td>-0.000347</td>\n",
       "      <td>0.004455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>dbnetis_at</td>\n",
       "      <td>0.021963</td>\n",
       "      <td>-0.000333</td>\n",
       "      <td>0.004278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>rmax1_21d</td>\n",
       "      <td>0.021883</td>\n",
       "      <td>-0.000253</td>\n",
       "      <td>0.003256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>ret_1_0</td>\n",
       "      <td>0.021793</td>\n",
       "      <td>-0.000163</td>\n",
       "      <td>0.002098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>be_me</td>\n",
       "      <td>0.021791</td>\n",
       "      <td>-0.000161</td>\n",
       "      <td>0.002067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>ni_be</td>\n",
       "      <td>0.021746</td>\n",
       "      <td>-0.000116</td>\n",
       "      <td>0.001485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>taccruals_ni</td>\n",
       "      <td>0.021737</td>\n",
       "      <td>-0.000107</td>\n",
       "      <td>0.001380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>cowc_gr1a</td>\n",
       "      <td>0.021672</td>\n",
       "      <td>-0.000042</td>\n",
       "      <td>0.000537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ni_me</td>\n",
       "      <td>0.021649</td>\n",
       "      <td>-0.000019</td>\n",
       "      <td>0.000247</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>ivol_capm_252d</td>\n",
       "      <td>0.021609</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>-0.000276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>debt_me</td>\n",
       "      <td>0.021555</td>\n",
       "      <td>0.000075</td>\n",
       "      <td>-0.000959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>cash_gr1a</td>\n",
       "      <td>0.021529</td>\n",
       "      <td>0.000101</td>\n",
       "      <td>-0.001295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>ret_12_7</td>\n",
       "      <td>0.021515</td>\n",
       "      <td>0.000115</td>\n",
       "      <td>-0.001478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>me</td>\n",
       "      <td>0.021414</td>\n",
       "      <td>0.000216</td>\n",
       "      <td>-0.002782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>be_gr1a</td>\n",
       "      <td>0.021402</td>\n",
       "      <td>0.000228</td>\n",
       "      <td>-0.002934</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>lnoa_gr1a</td>\n",
       "      <td>0.021331</td>\n",
       "      <td>0.000299</td>\n",
       "      <td>-0.003840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>rvol_21d</td>\n",
       "      <td>0.021226</td>\n",
       "      <td>0.000404</td>\n",
       "      <td>-0.005191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>ope_be</td>\n",
       "      <td>0.021187</td>\n",
       "      <td>0.000443</td>\n",
       "      <td>-0.005694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>inv_gr1a</td>\n",
       "      <td>0.021158</td>\n",
       "      <td>0.000472</td>\n",
       "      <td>-0.006072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>ret_6_0</td>\n",
       "      <td>0.021158</td>\n",
       "      <td>0.000472</td>\n",
       "      <td>-0.006073</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>ocf_at</td>\n",
       "      <td>0.021079</td>\n",
       "      <td>0.000551</td>\n",
       "      <td>-0.007080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>sale_gr1</td>\n",
       "      <td>0.021000</td>\n",
       "      <td>0.000630</td>\n",
       "      <td>-0.008095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>sale_gr3</td>\n",
       "      <td>0.020981</td>\n",
       "      <td>0.000649</td>\n",
       "      <td>-0.008348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>oaccruals_at</td>\n",
       "      <td>0.020846</td>\n",
       "      <td>0.000784</td>\n",
       "      <td>-0.010077</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>tax_gr1a</td>\n",
       "      <td>0.020801</td>\n",
       "      <td>0.000829</td>\n",
       "      <td>-0.010661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>taccruals_at</td>\n",
       "      <td>0.020702</td>\n",
       "      <td>0.000928</td>\n",
       "      <td>-0.011933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>gp_bev</td>\n",
       "      <td>0.020680</td>\n",
       "      <td>0.000950</td>\n",
       "      <td>-0.012207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ret_12_1</td>\n",
       "      <td>0.020632</td>\n",
       "      <td>0.000998</td>\n",
       "      <td>-0.012833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>chcsho_12m</td>\n",
       "      <td>0.019391</td>\n",
       "      <td>0.002240</td>\n",
       "      <td>-0.028788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>age</td>\n",
       "      <td>0.015604</td>\n",
       "      <td>0.006026</td>\n",
       "      <td>-0.077464</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Feature     R2OOS  red_R2OOS   var_imp\n",
       "45  prc_highprc_252d  0.036779  -0.015149  0.194736\n",
       "48        netdebt_me  0.029383  -0.007752  0.099655\n",
       "19          beta_60m  0.029104  -0.007474  0.096073\n",
       "9            sale_me  0.028714  -0.007084  0.091063\n",
       "27            at_gr1  0.028665  -0.007035  0.090431\n",
       "32       debtlt_gr1a  0.028607  -0.006977  0.089683\n",
       "17          sale_bev  0.028501  -0.006871  0.088324\n",
       "31          debt_gr1  0.028387  -0.006757  0.086856\n",
       "44          cop_atl1  0.028252  -0.006622  0.085125\n",
       "35        dsale_drec  0.028190  -0.006560  0.084330\n",
       "22           ret_9_1  0.027865  -0.006234  0.080141\n",
       "36           f_score  0.027851  -0.006221  0.079974\n",
       "21      bidaskhl_21d  0.022154  -0.000524  0.006736\n",
       "43            noa_at  0.022133  -0.000503  0.006466\n",
       "1            ret_6_1  0.022033  -0.000403  0.005181\n",
       "40            pi_nix  0.022025  -0.000395  0.005074\n",
       "16         ebit_sale  0.022013  -0.000383  0.004929\n",
       "25          ocf_debt  0.022013  -0.000383  0.004918\n",
       "34          caliq_cl  0.021987  -0.000357  0.004585\n",
       "29             ca_cl  0.021977  -0.000347  0.004455\n",
       "47        dbnetis_at  0.021963  -0.000333  0.004278\n",
       "20         rmax1_21d  0.021883  -0.000253  0.003256\n",
       "42           ret_1_0  0.021793  -0.000163  0.002098\n",
       "4              be_me  0.021791  -0.000161  0.002067\n",
       "24             ni_be  0.021746  -0.000116  0.001485\n",
       "15      taccruals_ni  0.021737  -0.000107  0.001380\n",
       "39         cowc_gr1a  0.021672  -0.000042  0.000537\n",
       "6              ni_me  0.021649  -0.000019  0.000247\n",
       "30    ivol_capm_252d  0.021609   0.000021 -0.000276\n",
       "5            debt_me  0.021555   0.000075 -0.000959\n",
       "26         cash_gr1a  0.021529   0.000101 -0.001295\n",
       "23          ret_12_7  0.021515   0.000115 -0.001478\n",
       "0                 me  0.021414   0.000216 -0.002782\n",
       "14           be_gr1a  0.021402   0.000228 -0.002934\n",
       "10         lnoa_gr1a  0.021331   0.000299 -0.003840\n",
       "37          rvol_21d  0.021226   0.000404 -0.005191\n",
       "33            ope_be  0.021187   0.000443 -0.005694\n",
       "11          inv_gr1a  0.021158   0.000472 -0.006072\n",
       "41           ret_6_0  0.021158   0.000472 -0.006073\n",
       "46            ocf_at  0.021079   0.000551 -0.007080\n",
       "8           sale_gr1  0.021000   0.000630 -0.008095\n",
       "7           sale_gr3  0.020981   0.000649 -0.008348\n",
       "12      oaccruals_at  0.020846   0.000784 -0.010077\n",
       "3           tax_gr1a  0.020801   0.000829 -0.010661\n",
       "13      taccruals_at  0.020702   0.000928 -0.011933\n",
       "38            gp_bev  0.020680   0.000950 -0.012207\n",
       "2           ret_12_1  0.020632   0.000998 -0.012833\n",
       "28        chcsho_12m  0.019391   0.002240 -0.028788\n",
       "18               age  0.015604   0.006026 -0.077464"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Convert dictionary to dataframe\n",
    "pd.DataFrame(dic.items())\n",
    "imp=pd.DataFrame(dic.items(), columns=['Feature', 'R2OOS'])\n",
    "# Feature: name of the variable whose values are set to zero\n",
    "imp[\"Feature\"] = imp[\"Feature\"].str[6:]\n",
    "\n",
    "# Calculate reduction in predictive R2OOS \n",
    "imp[\"red_R2OOS\"] = R2OOS_all -imp[\"R2OOS\"]\n",
    "imp[\"var_imp\"] = imp[\"red_R2OOS\"]/sum(imp[\"red_R2OOS\"])\n",
    "imp=imp.sort_values(by = ['var_imp'], ascending = False)\n",
    "imp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot variable importance measures for the top-20 most influential variables\n",
    "\n",
    "fea_imp_graph = imp.sort_values(['var_imp', 'Feature'], ascending=[True, False]).iloc[-20:]\n",
    "_ = fea_imp_graph.plot(kind='barh', x='Feature', y='var_imp', figsize=(20, 10))\n",
    "plt.title('FR')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "var_imp_rf=imp[[\"Feature\", \"var_imp\"]]\n",
    "var_imp_rf.to_csv(r'C:\\Users\\krist\\Documents\\Data\\variable importance\\var_imp_rf.csv', index = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Machine learning portfolios\n",
    "\n",
    "Explore whether predictability translates into portfolio gains"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>yhat</th>\n",
       "      <th>y_true</th>\n",
       "      <th>date</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MonthYear</th>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2010-01</th>\n",
       "      <th>comp_001166_02W</th>\n",
       "      <td>0.007341</td>\n",
       "      <td>0.035512</td>\n",
       "      <td>2010-01-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_001661_01W</th>\n",
       "      <td>0.007651</td>\n",
       "      <td>-0.041371</td>\n",
       "      <td>2010-01-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_002410_04W</th>\n",
       "      <td>0.003303</td>\n",
       "      <td>-0.053836</td>\n",
       "      <td>2010-01-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_002597_02W</th>\n",
       "      <td>0.002349</td>\n",
       "      <td>0.001862</td>\n",
       "      <td>2010-01-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_003820_01W</th>\n",
       "      <td>-0.007226</td>\n",
       "      <td>0.078967</td>\n",
       "      <td>2010-01-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2019-12</th>\n",
       "      <th>comp_330716_01W</th>\n",
       "      <td>0.004274</td>\n",
       "      <td>-0.012904</td>\n",
       "      <td>2019-12-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_331115_01W</th>\n",
       "      <td>0.005865</td>\n",
       "      <td>0.005400</td>\n",
       "      <td>2019-12-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_332311_01W</th>\n",
       "      <td>0.000225</td>\n",
       "      <td>-0.051340</td>\n",
       "      <td>2019-12-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_333885_01W</th>\n",
       "      <td>0.009342</td>\n",
       "      <td>0.003528</td>\n",
       "      <td>2019-12-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_334036_02W</th>\n",
       "      <td>-0.003514</td>\n",
       "      <td>-0.100876</td>\n",
       "      <td>2019-12-31</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>93417 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                               yhat    y_true       date\n",
       "MonthYear id                                            \n",
       "2010-01   comp_001166_02W  0.007341  0.035512 2010-01-31\n",
       "          comp_001661_01W  0.007651 -0.041371 2010-01-31\n",
       "          comp_002410_04W  0.003303 -0.053836 2010-01-31\n",
       "          comp_002597_02W  0.002349  0.001862 2010-01-31\n",
       "          comp_003820_01W -0.007226  0.078967 2010-01-31\n",
       "...                             ...       ...        ...\n",
       "2019-12   comp_330716_01W  0.004274 -0.012904 2019-12-31\n",
       "          comp_331115_01W  0.005865  0.005400 2019-12-31\n",
       "          comp_332311_01W  0.000225 -0.051340 2019-12-31\n",
       "          comp_333885_01W  0.009342  0.003528 2019-12-31\n",
       "          comp_334036_02W -0.003514 -0.100876 2019-12-31\n",
       "\n",
       "[93417 rows x 3 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Generate a results dataframe containing the model predictions (yhat) and the true values (y_true) \n",
    "#of the dependent variable for each stock in each month\n",
    "\n",
    "yhat = predictions_all.tolist()\n",
    "y_true = y_test_list_all.tolist()\n",
    "i = dates_all.tolist()\n",
    "\n",
    "results = pd.DataFrame(\n",
    "    {'identifier': i,\n",
    "     'yhat': yhat,\n",
    "     'y_true': y_true\n",
    "    })\n",
    "\n",
    "results[\"identifier\"]= results[\"identifier\"].astype(\"str\")\n",
    "results[\"date\"] = results[\"identifier\"].str[12:22]\n",
    "results[\"id\"] = results[\"identifier\"].str[36:51]\n",
    "results.drop([\"identifier\"],axis = 1, inplace=True)\n",
    "results['date'] = pd.to_datetime(results['date'], format='%Y-%m-%d')\n",
    "results['MonthYear'] = results['date'].dt.to_period('M')\n",
    "results = results.sort_values(by = ['date', 'id'], ascending = True)\n",
    "results = results.set_index(['MonthYear','id'])\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save results to use for Diebold-Mariano test\n",
    "rf= results.reset_index()\n",
    "rf.to_csv(r'C:\\Users\\krist\\Documents\\Data\\Predictions\\rf.csv', index = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>me2</th>\n",
       "      <th>ret</th>\n",
       "      <th>ret_exc</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MonthYear</th>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2003-01</th>\n",
       "      <th>comp_001166_02W</th>\n",
       "      <td>567.074646</td>\n",
       "      <td>-0.018948</td>\n",
       "      <td>-0.019948</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_001661_01W</th>\n",
       "      <td>5221.931641</td>\n",
       "      <td>0.005815</td>\n",
       "      <td>0.004815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_004367_02W</th>\n",
       "      <td>4139.599121</td>\n",
       "      <td>-0.103801</td>\n",
       "      <td>-0.104801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_004925_02W</th>\n",
       "      <td>15302.482422</td>\n",
       "      <td>-0.070425</td>\n",
       "      <td>-0.071425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_005959_01W</th>\n",
       "      <td>6355.691406</td>\n",
       "      <td>-0.085813</td>\n",
       "      <td>-0.086813</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2020-12</th>\n",
       "      <th>comp_333885_01W</th>\n",
       "      <td>10829.783203</td>\n",
       "      <td>0.128645</td>\n",
       "      <td>0.128545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_334302_02W</th>\n",
       "      <td>297.482697</td>\n",
       "      <td>0.097961</td>\n",
       "      <td>0.097861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_335706_02W</th>\n",
       "      <td>196.243454</td>\n",
       "      <td>0.114381</td>\n",
       "      <td>0.114281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_340115_01W</th>\n",
       "      <td>1799.958496</td>\n",
       "      <td>0.155813</td>\n",
       "      <td>0.155713</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comp_340153_01W</th>\n",
       "      <td>26788.144531</td>\n",
       "      <td>0.238817</td>\n",
       "      <td>0.238717</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>156948 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                    me2       ret   ret_exc\n",
       "MonthYear id                                               \n",
       "2003-01   comp_001166_02W    567.074646 -0.018948 -0.019948\n",
       "          comp_001661_01W   5221.931641  0.005815  0.004815\n",
       "          comp_004367_02W   4139.599121 -0.103801 -0.104801\n",
       "          comp_004925_02W  15302.482422 -0.070425 -0.071425\n",
       "          comp_005959_01W   6355.691406 -0.085813 -0.086813\n",
       "...                                 ...       ...       ...\n",
       "2020-12   comp_333885_01W  10829.783203  0.128645  0.128545\n",
       "          comp_334302_02W    297.482697  0.097961  0.097861\n",
       "          comp_335706_02W    196.243454  0.114381  0.114281\n",
       "          comp_340115_01W   1799.958496  0.155813  0.155713\n",
       "          comp_340153_01W  26788.144531  0.238817  0.238717\n",
       "\n",
       "[156948 rows x 3 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Import the original \"me\" avriable (before rank standartization) to use for the value weighting scheme\n",
    "\n",
    "unscaled_data = df[[\"id\", 'eom', \"me2\",\"ret\" ,\"ret_exc\"]].copy()\n",
    "unscaled_data['MonthYear'] = unscaled_data['eom'].dt.to_period('M')\n",
    "unscaled_data.drop('eom', axis=1, inplace=True)\n",
    "unscaled_data = unscaled_data.set_index(['MonthYear','id'])\n",
    "unscaled_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>prc</th>\n",
       "      <th>prc_t+1</th>\n",
       "      <th>ret_t+1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MonthYear</th>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2003-01</th>\n",
       "      <th>comp_001166_02W</th>\n",
       "      <td>11.486148</td>\n",
       "      <td>10.468435</td>\n",
       "      <td>-0.088604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003-02</th>\n",
       "      <th>comp_001166_02W</th>\n",
       "      <td>10.468435</td>\n",
       "      <td>10.107208</td>\n",
       "      <td>-0.034506</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003-03</th>\n",
       "      <th>comp_001166_02W</th>\n",
       "      <td>10.107208</td>\n",
       "      <td>13.610050</td>\n",
       "      <td>0.346569</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003-04</th>\n",
       "      <th>comp_001166_02W</th>\n",
       "      <td>13.610050</td>\n",
       "      <td>14.518828</td>\n",
       "      <td>0.066773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003-05</th>\n",
       "      <th>comp_001166_02W</th>\n",
       "      <td>14.518828</td>\n",
       "      <td>15.212792</td>\n",
       "      <td>0.047798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-11</th>\n",
       "      <th>comp_340115_01W</th>\n",
       "      <td>14.831518</td>\n",
       "      <td>17.142462</td>\n",
       "      <td>0.155813</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-12</th>\n",
       "      <th>comp_340115_01W</th>\n",
       "      <td>17.142462</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-10</th>\n",
       "      <th>comp_340153_01W</th>\n",
       "      <td>21.909940</td>\n",
       "      <td>29.758645</td>\n",
       "      <td>0.358226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-11</th>\n",
       "      <th>comp_340153_01W</th>\n",
       "      <td>29.758645</td>\n",
       "      <td>36.865510</td>\n",
       "      <td>0.238817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-12</th>\n",
       "      <th>comp_340153_01W</th>\n",
       "      <td>36.865510</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>274031 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 prc    prc_t+1   ret_t+1\n",
       "MonthYear id                                             \n",
       "2003-01   comp_001166_02W  11.486148  10.468435 -0.088604\n",
       "2003-02   comp_001166_02W  10.468435  10.107208 -0.034506\n",
       "2003-03   comp_001166_02W  10.107208  13.610050  0.346569\n",
       "2003-04   comp_001166_02W  13.610050  14.518828  0.066773\n",
       "2003-05   comp_001166_02W  14.518828  15.212792  0.047798\n",
       "...                              ...        ...       ...\n",
       "2020-11   comp_340115_01W  14.831518  17.142462  0.155813\n",
       "2020-12   comp_340115_01W  17.142462        NaN       NaN\n",
       "2020-10   comp_340153_01W  21.909940  29.758645  0.358226\n",
       "2020-11   comp_340153_01W  29.758645  36.865510  0.238817\n",
       "2020-12   comp_340153_01W  36.865510        NaN       NaN\n",
       "\n",
       "[274031 rows x 3 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Import price data in t+1 and ret t+1 --> Import from the original dataframe\n",
    "# (otherwise there are missing values because shares with a price < $5 have been excluded)\n",
    "\n",
    "price_data= pd.read_csv(r\"C:\\Users\\krist\\Documents\\Data\\ger_factor_data.csv\", dtype ={\"comp_tpci\": str}, parse_dates =[\"eom\"])\n",
    "price_data[\"eom\"] = pd.to_datetime(price_data[\"eom\"], utc = False)\n",
    "\n",
    "# Only the data between 2003 and 2020 is needed\n",
    "price_data[\"year\"] = price_data[\"eom\"].dt.year\n",
    "price_data = price_data.loc[(price_data[\"year\"]>=2003) & (price_data[\"year\"]<=2020)]\n",
    "price_data = price_data[[\"id\", 'eom', 'prc', \"ret\"]]\n",
    "\n",
    "#Generate a MonthYear variable (monthly observations are given, the day is not needed)\n",
    "price_data['MonthYear'] = price_data['eom'].dt.to_period('M')\n",
    "price_data.drop('eom', axis=1, inplace=True)\n",
    "\n",
    "#The original dataframe is not winsorized --> winsorize  at the 1% and 99% level \n",
    "price = [\"prc\"]\n",
    "ret = [\"ret\"]\n",
    "price_data[\"prc\"] = price_data[price].apply(lambda x: x.clip(*x.quantile([0.01, 0.99])))\n",
    "price_data[\"ret\"] = price_data[ret].apply(lambda x: x.clip(*x.quantile([0.01, 0.99])))\n",
    "\n",
    "#Shift to get the price of each stock in month t+1\n",
    "price_data['prc_t+1'] = price_data.groupby(['id'])['prc'].shift(-1)\n",
    "# Shift to get return for each stock in month t+1\n",
    "price_data[\"ret_t+1\"]=price_data.groupby(['id'])['ret'].shift(-1)\n",
    "price_data.drop('ret', axis=1, inplace=True)\n",
    "price_data = price_data.set_index(['MonthYear','id'])\n",
    "price_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MonthYear</th>\n",
       "      <th>id</th>\n",
       "      <th>yhat</th>\n",
       "      <th>y_true</th>\n",
       "      <th>date</th>\n",
       "      <th>me</th>\n",
       "      <th>ret</th>\n",
       "      <th>ret_exc</th>\n",
       "      <th>prc</th>\n",
       "      <th>prc_t+1</th>\n",
       "      <th>ret_t+1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2010-01</td>\n",
       "      <td>comp_001166_02W</td>\n",
       "      <td>0.007341</td>\n",
       "      <td>0.035512</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>1233.673950</td>\n",
       "      <td>-0.099572</td>\n",
       "      <td>-0.099572</td>\n",
       "      <td>22.730003</td>\n",
       "      <td>23.537181</td>\n",
       "      <td>0.035512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2010-01</td>\n",
       "      <td>comp_001661_01W</td>\n",
       "      <td>0.007651</td>\n",
       "      <td>-0.041371</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>6796.776367</td>\n",
       "      <td>0.036020</td>\n",
       "      <td>0.036020</td>\n",
       "      <td>22.730003</td>\n",
       "      <td>21.789641</td>\n",
       "      <td>-0.041371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2010-01</td>\n",
       "      <td>comp_002410_04W</td>\n",
       "      <td>0.003303</td>\n",
       "      <td>-0.053836</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>132047.843750</td>\n",
       "      <td>-0.014941</td>\n",
       "      <td>-0.014941</td>\n",
       "      <td>9.453457</td>\n",
       "      <td>8.812790</td>\n",
       "      <td>-0.053836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2010-01</td>\n",
       "      <td>comp_002597_02W</td>\n",
       "      <td>0.002349</td>\n",
       "      <td>0.001862</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>18138.388672</td>\n",
       "      <td>0.035186</td>\n",
       "      <td>0.035186</td>\n",
       "      <td>13.191743</td>\n",
       "      <td>13.057401</td>\n",
       "      <td>0.001862</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2010-01</td>\n",
       "      <td>comp_003820_01W</td>\n",
       "      <td>-0.007226</td>\n",
       "      <td>0.078967</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>225.897369</td>\n",
       "      <td>-0.125756</td>\n",
       "      <td>-0.125756</td>\n",
       "      <td>6.339377</td>\n",
       "      <td>6.839981</td>\n",
       "      <td>0.078967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93412</th>\n",
       "      <td>2019-12</td>\n",
       "      <td>comp_330716_01W</td>\n",
       "      <td>0.004274</td>\n",
       "      <td>-0.012904</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>33.930836</td>\n",
       "      <td>0.066508</td>\n",
       "      <td>0.065108</td>\n",
       "      <td>6.169243</td>\n",
       "      <td>6.097655</td>\n",
       "      <td>-0.011604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93413</th>\n",
       "      <td>2019-12</td>\n",
       "      <td>comp_331115_01W</td>\n",
       "      <td>0.005865</td>\n",
       "      <td>0.005400</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>36.342445</td>\n",
       "      <td>0.005615</td>\n",
       "      <td>0.004215</td>\n",
       "      <td>18.171224</td>\n",
       "      <td>18.292966</td>\n",
       "      <td>0.006700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93414</th>\n",
       "      <td>2019-12</td>\n",
       "      <td>comp_332311_01W</td>\n",
       "      <td>0.000225</td>\n",
       "      <td>-0.051340</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>13412.494141</td>\n",
       "      <td>-0.020668</td>\n",
       "      <td>-0.022068</td>\n",
       "      <td>26.824988</td>\n",
       "      <td>25.482656</td>\n",
       "      <td>-0.050040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93415</th>\n",
       "      <td>2019-12</td>\n",
       "      <td>comp_333885_01W</td>\n",
       "      <td>0.009342</td>\n",
       "      <td>0.003528</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>7151.834473</td>\n",
       "      <td>0.251631</td>\n",
       "      <td>0.250231</td>\n",
       "      <td>35.759173</td>\n",
       "      <td>35.931819</td>\n",
       "      <td>0.004828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93416</th>\n",
       "      <td>2019-12</td>\n",
       "      <td>comp_334036_02W</td>\n",
       "      <td>-0.003514</td>\n",
       "      <td>-0.100876</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>189.330505</td>\n",
       "      <td>0.377943</td>\n",
       "      <td>0.377359</td>\n",
       "      <td>21.424097</td>\n",
       "      <td>19.290764</td>\n",
       "      <td>-0.099576</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>93417 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      MonthYear               id      yhat    y_true       date  \\\n",
       "0       2010-01  comp_001166_02W  0.007341  0.035512 2010-01-31   \n",
       "1       2010-01  comp_001661_01W  0.007651 -0.041371 2010-01-31   \n",
       "2       2010-01  comp_002410_04W  0.003303 -0.053836 2010-01-31   \n",
       "3       2010-01  comp_002597_02W  0.002349  0.001862 2010-01-31   \n",
       "4       2010-01  comp_003820_01W -0.007226  0.078967 2010-01-31   \n",
       "...         ...              ...       ...       ...        ...   \n",
       "93412   2019-12  comp_330716_01W  0.004274 -0.012904 2019-12-31   \n",
       "93413   2019-12  comp_331115_01W  0.005865  0.005400 2019-12-31   \n",
       "93414   2019-12  comp_332311_01W  0.000225 -0.051340 2019-12-31   \n",
       "93415   2019-12  comp_333885_01W  0.009342  0.003528 2019-12-31   \n",
       "93416   2019-12  comp_334036_02W -0.003514 -0.100876 2019-12-31   \n",
       "\n",
       "                  me       ret   ret_exc        prc    prc_t+1   ret_t+1  \n",
       "0        1233.673950 -0.099572 -0.099572  22.730003  23.537181  0.035512  \n",
       "1        6796.776367  0.036020  0.036020  22.730003  21.789641 -0.041371  \n",
       "2      132047.843750 -0.014941 -0.014941   9.453457   8.812790 -0.053836  \n",
       "3       18138.388672  0.035186  0.035186  13.191743  13.057401  0.001862  \n",
       "4         225.897369 -0.125756 -0.125756   6.339377   6.839981  0.078967  \n",
       "...              ...       ...       ...        ...        ...       ...  \n",
       "93412      33.930836  0.066508  0.065108   6.169243   6.097655 -0.011604  \n",
       "93413      36.342445  0.005615  0.004215  18.171224  18.292966  0.006700  \n",
       "93414   13412.494141 -0.020668 -0.022068  26.824988  25.482656 -0.050040  \n",
       "93415    7151.834473  0.251631  0.250231  35.759173  35.931819  0.004828  \n",
       "93416     189.330505  0.377943  0.377359  21.424097  19.290764 -0.099576  \n",
       "\n",
       "[93417 rows x 11 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Merge with the results dataframe\n",
    "bigdata = pd.merge(results, unscaled_data,left_index=True, right_index=True)\n",
    "bigdata = pd.merge(bigdata, price_data,left_index=True, right_index=True)\n",
    "bigdata = bigdata.reset_index()\n",
    "\n",
    "# Original \"me\" (before rank normalization) to use for the value weighting scheme\n",
    "bigdata.rename(columns={'me2': 'me'}, inplace=True)\n",
    "\n",
    "#Final df containing all information needed for the portfolio creation\n",
    "bigdata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MonthYear     0\n",
       "id            0\n",
       "yhat          0\n",
       "y_true        0\n",
       "date          0\n",
       "me            0\n",
       "ret           0\n",
       "ret_exc       0\n",
       "prc           0\n",
       "prc_t+1      12\n",
       "ret_t+1      12\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Check for missing values (the original df has not yet been cleaned of missing values)\n",
    "bigdata.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "93405"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Drop the twelve observations where prc_t+1 and ret_t+1 are still missing\n",
    "bigdata = bigdata.dropna()\n",
    "len(bigdata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate the 1-month risk-free rate as the average monthly difference between returns and excess returns\n",
    "bigdata[\"risk_free_rate\"]=bigdata[\"ret_t+1\"]-bigdata[\"y_true\"]\n",
    "bigdata[\"risk_free_rate\"] =bigdata.groupby('MonthYear')[\"risk_free_rate\"].transform('mean')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  1,   2,   3,   4,   5,   6,   7,   8,   9,  10,  11,  12,  13,\n",
       "        14,  15,  16,  17,  18,  19,  20,  21,  22,  23,  24,  25,  26,\n",
       "        27,  28,  29,  30,  31,  32,  33,  34,  35,  36,  37,  38,  39,\n",
       "        40,  41,  42,  43,  44,  45,  46,  47,  48,  49,  50,  51,  52,\n",
       "        53,  54,  55,  56,  57,  58,  59,  60,  61,  62,  63,  64,  65,\n",
       "        66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,\n",
       "        79,  80,  81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,\n",
       "        92,  93,  94,  95,  96,  97,  98,  99, 100, 101, 102, 103, 104,\n",
       "       105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117,\n",
       "       118, 119, 120], dtype=int64)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a variable \"NumMonth\", indicating the number of the respectie month: \n",
    "# January 2010 = 1,....,December 2019 = 120\n",
    "\n",
    "bigdata[\"MonthYear1\"]= bigdata[\"MonthYear\"].copy()\n",
    "bigdata[\"MonthYear\"]= bigdata[\"MonthYear\"].astype(\"int\")\n",
    "bigdata[\"NumMonth\"]= bigdata[\"MonthYear\"]-479\n",
    "bigdata[\"NumMonth\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a separate dataframe for each month,e.g.: \n",
    "# df_1=January 2010, df_2 = February 2010,....,df_120 = December 2019\n",
    "\n",
    "for i in bigdata[\"NumMonth\"].unique():\n",
    "    globals()['df_' + str(i)] =  bigdata[bigdata[\"NumMonth\"]==i]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Assign decile ranks to the stocks within each month based on the predicted excess returns:\n",
    "\n",
    "# 1)Generate ranks starting from the lowest predicted excess return to the highest\n",
    "for i in bigdata[\"NumMonth\"].unique():\n",
    "    globals()['df_' + str(i)][\"rank\"]= globals()['df_' + str(i)]['yhat'].rank(method='first')\n",
    "    \n",
    "# 2)Split ranks into 10 deciles to get decile rank\n",
    "# DecileRank = 0: stoks with lowest predicted excess returns\n",
    "# DecileRank =9: stoks with highest predicted excess returns\n",
    "for i in bigdata[\"NumMonth\"].unique():\n",
    "    globals()['df_' + str(i)][\"DecileRank\"]=pd.qcut(globals()['df_' + str(i)]['rank'].values, 10, labels = False)\n",
    "\n",
    "#Drop normal rank, retain only decile ranks \n",
    "for i in bigdata[\"NumMonth\"].unique():\n",
    "     globals()['df_' + str(i)].drop('rank', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a separate dataframe for each decile rank and each month,e.g.: \n",
    "# df_1_0: the bottom decile in January 2010,\n",
    "# d_1_9: the top decile in January 2010,....,\n",
    "# df_120_0: the bottom decile in December 2019,\n",
    "# df_120_9: the top decile in December 2019\n",
    "\n",
    "for i in bigdata[\"NumMonth\"].unique():\n",
    "    for j,g in globals()['df_' + str(i)].groupby('DecileRank'):\n",
    "        globals()['df_' + str(i)+ \"_\" + str(j)] =  g"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# All stocks are sorted into deciles based on their predicted returns for the next month. \n",
    "# Create 10 dataframes showing the portfolio composition of each decile portfolio over time\n",
    "\n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['rank_' + str(j)] = pd.concat([globals()['df_1_'+ str(j)], globals()['df_2_'+ str(j)]], axis=0)\n",
    "    \n",
    "# Generate 10 Dataframes for the 10 Decile portfolios 0-9: rank_9: top portfolio, rank_0: bottom portfolio\n",
    "for i in np.arange(2,120,1):\n",
    "    for j in np.arange(0,10,1):\n",
    "        globals()['rank_' + str(j)] = pd.concat([globals()['rank_' + str(j)], globals()['df_' + str(i+1)+ \"_\" + str(j)]], axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MonthYear</th>\n",
       "      <th>id</th>\n",
       "      <th>yhat</th>\n",
       "      <th>y_true</th>\n",
       "      <th>date</th>\n",
       "      <th>me</th>\n",
       "      <th>ret</th>\n",
       "      <th>ret_exc</th>\n",
       "      <th>prc</th>\n",
       "      <th>prc_t+1</th>\n",
       "      <th>ret_t+1</th>\n",
       "      <th>risk_free_rate</th>\n",
       "      <th>MonthYear1</th>\n",
       "      <th>NumMonth</th>\n",
       "      <th>DecileRank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>480</td>\n",
       "      <td>comp_011749_09W</td>\n",
       "      <td>0.019561</td>\n",
       "      <td>-0.013695</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>6580.630859</td>\n",
       "      <td>-0.041260</td>\n",
       "      <td>-0.041260</td>\n",
       "      <td>16.056975</td>\n",
       "      <td>15.837082</td>\n",
       "      <td>-0.013695</td>\n",
       "      <td>0.000353</td>\n",
       "      <td>2010-01</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>480</td>\n",
       "      <td>comp_014269_01W</td>\n",
       "      <td>0.019927</td>\n",
       "      <td>-0.034796</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>2671.458984</td>\n",
       "      <td>0.006921</td>\n",
       "      <td>0.006921</td>\n",
       "      <td>40.510844</td>\n",
       "      <td>39.101210</td>\n",
       "      <td>-0.034796</td>\n",
       "      <td>0.000353</td>\n",
       "      <td>2010-01</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>480</td>\n",
       "      <td>comp_015321_02W</td>\n",
       "      <td>0.019476</td>\n",
       "      <td>-0.053604</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>32227.082031</td>\n",
       "      <td>-0.079324</td>\n",
       "      <td>-0.079324</td>\n",
       "      <td>30.438742</td>\n",
       "      <td>28.807106</td>\n",
       "      <td>-0.053604</td>\n",
       "      <td>0.000353</td>\n",
       "      <td>2010-01</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>480</td>\n",
       "      <td>comp_017420_03W</td>\n",
       "      <td>0.020917</td>\n",
       "      <td>0.059737</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>11078.906250</td>\n",
       "      <td>-0.107508</td>\n",
       "      <td>-0.107508</td>\n",
       "      <td>13.198694</td>\n",
       "      <td>13.987147</td>\n",
       "      <td>0.059737</td>\n",
       "      <td>0.000353</td>\n",
       "      <td>2010-01</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>480</td>\n",
       "      <td>comp_019565_13W</td>\n",
       "      <td>0.021830</td>\n",
       "      <td>-0.007516</td>\n",
       "      <td>2010-01-31</td>\n",
       "      <td>80014.750000</td>\n",
       "      <td>-0.030124</td>\n",
       "      <td>-0.030124</td>\n",
       "      <td>57.794162</td>\n",
       "      <td>57.359775</td>\n",
       "      <td>-0.007516</td>\n",
       "      <td>0.000353</td>\n",
       "      <td>2010-01</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93404</th>\n",
       "      <td>599</td>\n",
       "      <td>comp_328809_02W</td>\n",
       "      <td>0.008740</td>\n",
       "      <td>-0.035295</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>2947.826660</td>\n",
       "      <td>0.263338</td>\n",
       "      <td>0.261938</td>\n",
       "      <td>34.659927</td>\n",
       "      <td>33.481671</td>\n",
       "      <td>-0.033995</td>\n",
       "      <td>0.001325</td>\n",
       "      <td>2019-12</td>\n",
       "      <td>120</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93409</th>\n",
       "      <td>599</td>\n",
       "      <td>comp_329320_01W</td>\n",
       "      <td>0.008423</td>\n",
       "      <td>0.140847</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>240.979858</td>\n",
       "      <td>0.190764</td>\n",
       "      <td>0.189364</td>\n",
       "      <td>18.171224</td>\n",
       "      <td>20.754201</td>\n",
       "      <td>0.142147</td>\n",
       "      <td>0.001325</td>\n",
       "      <td>2019-12</td>\n",
       "      <td>120</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93411</th>\n",
       "      <td>599</td>\n",
       "      <td>comp_330213_01W</td>\n",
       "      <td>0.009397</td>\n",
       "      <td>0.150887</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>703.357910</td>\n",
       "      <td>0.029799</td>\n",
       "      <td>0.028399</td>\n",
       "      <td>9.814704</td>\n",
       "      <td>11.308379</td>\n",
       "      <td>0.152187</td>\n",
       "      <td>0.001325</td>\n",
       "      <td>2019-12</td>\n",
       "      <td>120</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93413</th>\n",
       "      <td>599</td>\n",
       "      <td>comp_331115_01W</td>\n",
       "      <td>0.005865</td>\n",
       "      <td>0.005400</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>36.342445</td>\n",
       "      <td>0.005615</td>\n",
       "      <td>0.004215</td>\n",
       "      <td>18.171224</td>\n",
       "      <td>18.292966</td>\n",
       "      <td>0.006700</td>\n",
       "      <td>0.001325</td>\n",
       "      <td>2019-12</td>\n",
       "      <td>120</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93415</th>\n",
       "      <td>599</td>\n",
       "      <td>comp_333885_01W</td>\n",
       "      <td>0.009342</td>\n",
       "      <td>0.003528</td>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>7151.834473</td>\n",
       "      <td>0.251631</td>\n",
       "      <td>0.250231</td>\n",
       "      <td>35.759173</td>\n",
       "      <td>35.931819</td>\n",
       "      <td>0.004828</td>\n",
       "      <td>0.001325</td>\n",
       "      <td>2019-12</td>\n",
       "      <td>120</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>9386 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       MonthYear               id      yhat    y_true       date  \\\n",
       "16           480  comp_011749_09W  0.019561 -0.013695 2010-01-31   \n",
       "21           480  comp_014269_01W  0.019927 -0.034796 2010-01-31   \n",
       "26           480  comp_015321_02W  0.019476 -0.053604 2010-01-31   \n",
       "41           480  comp_017420_03W  0.020917  0.059737 2010-01-31   \n",
       "48           480  comp_019565_13W  0.021830 -0.007516 2010-01-31   \n",
       "...          ...              ...       ...       ...        ...   \n",
       "93404        599  comp_328809_02W  0.008740 -0.035295 2019-12-31   \n",
       "93409        599  comp_329320_01W  0.008423  0.140847 2019-12-31   \n",
       "93411        599  comp_330213_01W  0.009397  0.150887 2019-12-31   \n",
       "93413        599  comp_331115_01W  0.005865  0.005400 2019-12-31   \n",
       "93415        599  comp_333885_01W  0.009342  0.003528 2019-12-31   \n",
       "\n",
       "                 me       ret   ret_exc        prc    prc_t+1   ret_t+1  \\\n",
       "16      6580.630859 -0.041260 -0.041260  16.056975  15.837082 -0.013695   \n",
       "21      2671.458984  0.006921  0.006921  40.510844  39.101210 -0.034796   \n",
       "26     32227.082031 -0.079324 -0.079324  30.438742  28.807106 -0.053604   \n",
       "41     11078.906250 -0.107508 -0.107508  13.198694  13.987147  0.059737   \n",
       "48     80014.750000 -0.030124 -0.030124  57.794162  57.359775 -0.007516   \n",
       "...             ...       ...       ...        ...        ...       ...   \n",
       "93404   2947.826660  0.263338  0.261938  34.659927  33.481671 -0.033995   \n",
       "93409    240.979858  0.190764  0.189364  18.171224  20.754201  0.142147   \n",
       "93411    703.357910  0.029799  0.028399   9.814704  11.308379  0.152187   \n",
       "93413     36.342445  0.005615  0.004215  18.171224  18.292966  0.006700   \n",
       "93415   7151.834473  0.251631  0.250231  35.759173  35.931819  0.004828   \n",
       "\n",
       "       risk_free_rate MonthYear1  NumMonth  DecileRank  \n",
       "16           0.000353    2010-01         1           9  \n",
       "21           0.000353    2010-01         1           9  \n",
       "26           0.000353    2010-01         1           9  \n",
       "41           0.000353    2010-01         1           9  \n",
       "48           0.000353    2010-01         1           9  \n",
       "...               ...        ...       ...         ...  \n",
       "93404        0.001325    2019-12       120           9  \n",
       "93409        0.001325    2019-12       120           9  \n",
       "93411        0.001325    2019-12       120           9  \n",
       "93413        0.001325    2019-12       120           9  \n",
       "93415        0.001325    2019-12       120           9  \n",
       "\n",
       "[9386 rows x 15 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We get 10 separate dataframes (rank_0, rank_1,...., rank_9), showing the portfolio \n",
    "# composition of each decile portfolio over time\n",
    "#e.g.: compostion of the top decile portfolio over the entire OOS period (Jan 2010-Dec 2019):\n",
    "rank_9"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Portfolio performance\n",
    "\n",
    "Next, the performance of the prediction-sorted portfolios is calculated over the 10-year out-of-sample testing for both equal and value weighting schemes\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get equal und value weights per stock per month in each decile portfolio:\n",
    "# ew = Equally weighted\n",
    "# vw = Value weighted\n",
    "\n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['rank_' + str(j)][\"eq_weights\"] = 1/globals()['rank_' + str(j)].groupby('MonthYear')[\"id\"].transform('size')\n",
    "    globals()['rank_' + str(j)][\"me_weights\"] = globals()['rank_' + str(j)][\"me\"]/globals()['rank_' + str(j)].groupby('MonthYear')[\"me\"].transform('sum')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### The excess return of each decile portfolio in month t is modeled as the weighted sum of each stocks’s  excess return in month t:\n",
    "\n",
    "#### $$\n",
    "R_{P, t}=\\sum_{i=1}^{N} w_{i, t} * r_{i, t}\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Weighted excess return per stock in t+1\n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['rank_' + str(j)]['excess_return_stock_ew'] = globals()['rank_' + str(j)][\"y_true\"]*globals()['rank_' + str(j)][\"eq_weights\"]\n",
    "    globals()['rank_' + str(j)]['excess_return_stock_vw'] = globals()['rank_' + str(j)][\"y_true\"]*globals()['rank_' + str(j)][\"me_weights\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# weighted return per stock in t+1 (to use for the sharpe ratio)\n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['rank_' + str(j)]['return_stock_ew'] = globals()['rank_' + str(j)][\"ret_t+1\"]*globals()['rank_' + str(j)][\"eq_weights\"]\n",
    "    globals()['rank_' + str(j)]['return_stock_vw'] = globals()['rank_' + str(j)][\"ret_t+1\"]*globals()['rank_' + str(j)][\"me_weights\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Portfolio excess return in t+1\n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['rank_' + str(j)]['excess_return_portfolio_ew'] = globals()['rank_' + str(j)].groupby('MonthYear')[\"excess_return_stock_ew\"].transform('sum')\n",
    "    globals()['rank_' + str(j)]['excess_return_portfolio_vw'] = globals()['rank_' + str(j)].groupby('MonthYear')[\"excess_return_stock_vw\"].transform('sum')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Portfolio return in t+1 (to use for the sharpe ratio) \n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['rank_' + str(j)]['return_portfolio_ew'] = globals()['rank_' + str(j)].groupby('MonthYear')[\"return_stock_ew\"].transform('sum')\n",
    "    globals()['rank_' + str(j)]['return_portfolio_vw'] = globals()['rank_' + str(j)].groupby('MonthYear')[\"return_stock_vw\"].transform('sum')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Weighted predicted excess return per stock in t+1\n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['rank_' + str(j)]['pred_excess_return_stock_ew'] = globals()['rank_' + str(j)][\"yhat\"]*globals()['rank_' + str(j)][\"eq_weights\"]\n",
    "    globals()['rank_' + str(j)]['pred_excess_return_stock_vw'] = globals()['rank_' + str(j)][\"yhat\"]*globals()['rank_' + str(j)][\"me_weights\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Portfolio predicted excess return in t+1\n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['rank_' + str(j)]['pred_excess_return_portfolio_ew'] = globals()['rank_' + str(j)].groupby('MonthYear')[\"pred_excess_return_stock_ew\"].transform('sum')\n",
    "    globals()['rank_' + str(j)]['pred_excess_return_portfolio_vw'] = globals()['rank_' + str(j)].groupby('MonthYear')[\"pred_excess_return_stock_vw\"].transform('sum')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate dataframes, containing the portfolio returns on mohtly basis for each decile portfolio\n",
    "# e.g., montly_rank_0: dataframe, containing only the monthly portfolio excess returns (predicted and real) \n",
    "# for  the bottom rank\n",
    "\n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['montly_rank_' + str(j)] = globals()['rank_' + str(j)][[\"MonthYear1\", \"DecileRank\",\n",
    "                                                                      \"excess_return_portfolio_ew\",\n",
    "                                                                      \"excess_return_portfolio_vw\", \n",
    "                                                                      \"pred_excess_return_portfolio_ew\",\n",
    "                                                                      \"pred_excess_return_portfolio_vw\",\n",
    "                                                                      \"return_portfolio_ew\",\n",
    "                                                                      \"return_portfolio_vw\"]]\n",
    "    \n",
    "for j in np.arange(0,10,1):\n",
    "    globals()['montly_rank_' + str(j)]=globals()['montly_rank_' + str(j)].drop_duplicates()\n",
    "    globals()['montly_rank_' + str(j)]=globals()['montly_rank_' + str(j)].set_index(\"MonthYear1\")\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Calculate time-series averages of both predicted and realized excess returns for each decile portfolio\n",
    "- Calculate the standard deviations and the annualized sharpe ratios based on the realized excess returns\n",
    "- Get measures for both equal- and value-weighted  portfolios \n",
    "\n",
    "### $\\text { Sharpe Ratio }=\\frac{\\text { Portfolio return-Risk Free } \\text { Rate }}{\\text { Standard deviation of portfolio reurn }}=\\frac{R_{p}-R_{F}}{\\sigma_{P}}=\\frac{\\text { Portfolio excess return }}{\\text { Std(portfolio return) }}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "for j in np.arange(0,10,1):\n",
    "    #Time-series average of realized excess returns\n",
    "    globals()[\"ew_mean_return_rank_\" +  str(j)]= globals()['montly_rank_' + str(j)][\"excess_return_portfolio_ew\"].mean()\n",
    "    globals()[\"vw_mean_return_rank_\" +  str(j)]= globals()['montly_rank_' + str(j)][\"excess_return_portfolio_vw\"].mean()\n",
    "    #Time-series average of predicted excess returns\n",
    "    globals()[\"ew_mean_pred_return_rank_\" +  str(j)]= globals()['montly_rank_' + str(j)][\"pred_excess_return_portfolio_ew\"].mean()\n",
    "    globals()[\"vw_mean_pred_return_rank_\" +  str(j)]= globals()['montly_rank_' + str(j)][\"pred_excess_return_portfolio_vw\"].mean()\n",
    "    #Standard deviation of realized excess returns\n",
    "    globals()[\"std_ew_rank_\" +  str(j)]= globals()['montly_rank_' + str(j)][\"excess_return_portfolio_ew\"].std()\n",
    "    globals()[\"std_vw_rank_\" +  str(j)]= globals()['montly_rank_' + str(j)][\"excess_return_portfolio_vw\"].std()\n",
    "    #Annualized sharpe ratio of realized excess returns\n",
    "    globals()[\"sharpe_ew_rank_\" +  str(j)]= (globals()['montly_rank_' + str(j)][\"excess_return_portfolio_ew\"].mean()/globals()['montly_rank_' + str(j)][\"return_portfolio_ew\"].std())* np.sqrt(12)\n",
    "    globals()[\"sharpe_vw_rank_\" +  str(j)]= (globals()['montly_rank_' + str(j)][\"excess_return_portfolio_vw\"].mean()/globals()['montly_rank_' + str(j)][\"return_portfolio_vw\"].std())* np.sqrt(12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Zero net investment long-short portfolio\n",
    "\n",
    "\n",
    "A zero-investment portfolio is a portfolio that has a net value of zero when the portfolio is assembled, and therefore requires an investor to take no equity stake in the portfolio. \n",
    " - e.g. short sell **USD 1,000** worth of the bottom portfolio, and use the proceeds to purchase **USD 1,000** worth of the top portfolio \n",
    " - this results in a net value of zero\n",
    " \n",
    " \n",
    "#### $\\sum_{i=1}^{N} w_{i, \\text { long }}=1$\n",
    "#### $\\sum_{i=1}^{N} w_{i, \\text { short }}=-1$\n",
    "#### $\\sum_{i=1}^{N} w_{i, \\text { long }} *\\text { prc }_{i, \\text { long }}=\\sum_{i=1}^{N} w_{i, \\text { short }} *\\text { prc }_{i, \\text { short }}$\n",
    "#### $\\text { Value Zero Net Portfolio in } t=\\sum_{i=1}^{N} w_{i, \\text { long }} * \\text { prc }_{i, \\text { long }}-\\sum_{i=1}^{N} w_{i, \\text { short }} * \\text { prc }_{i, \\text { short }}=0$\n",
    "\n",
    "\n",
    "#### $\\text { Return Zero Net Portfolio in } t=\\sum_{i=1}^{N} w_{i, \\text { long }} * \\text { ret }_{i, \\text { long }}-\\sum_{i=1}^{N} w_{i, \\text { short }} * \\text { ret }_{i, \\text { short }}$\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# For the zero-net-investment long-short portfolio the top (long) and bottom(short) decile portfolios are needed\n",
    "\n",
    "long_monthly = rank_9[[\"NumMonth\",\"MonthYear1\", \"DecileRank\", \"excess_return_portfolio_ew\",\n",
    "                       \"excess_return_portfolio_vw\",\"pred_excess_return_portfolio_ew\",\n",
    "                       \"pred_excess_return_portfolio_vw\",\"return_portfolio_ew\",\n",
    "                        \"return_portfolio_vw\"]].drop_duplicates()\n",
    "\n",
    "short_monthly = rank_0[[\"NumMonth\",\"MonthYear1\", \"DecileRank\", \"excess_return_portfolio_ew\",\n",
    "                       \"excess_return_portfolio_vw\",\"pred_excess_return_portfolio_ew\",\n",
    "                       \"pred_excess_return_portfolio_vw\",\"return_portfolio_ew\",\n",
    "                        \"return_portfolio_vw\"]].drop_duplicates()\n",
    "\n",
    "# Create a column, indication the stategy \n",
    "long_monthly[\"Strategy\"]= \"long\"\n",
    "short_monthly[\"Strategy\"]= \"short\"\n",
    "\n",
    "# Merge to get the zero net investment portfolio\n",
    "zeronet_monthly= pd.concat([long_monthly, short_monthly])\n",
    "zeronet_monthly = zeronet_monthly.sort_values(by = ['NumMonth',\"Strategy\"])\n",
    "zeronet_monthly[\"return_portfolio_vw\"] = zeronet_monthly[\"return_portfolio_vw\"].astype('float64')\n",
    "\n",
    "#Create two new columns containing the exess return of the portfolio and initially set the values to zero.\n",
    "zeronet_monthly[\"excess_return_zeronet_ew\"] =0\n",
    "zeronet_monthly[\"excess_return_zeronet_vw\"] =0\n",
    "\n",
    "# excess return zeronet in t = (weigted excess return long in t) - (weigted excess return short in t)\n",
    "for i in range(0, len(zeronet_monthly)):\n",
    "    if zeronet_monthly.iloc[i,9] == \"long\":\n",
    "        zeronet_monthly.iloc[i, -2] = zeronet_monthly.iloc[i, 3]-zeronet_monthly.iloc[i+1, 3]\n",
    "    else:\n",
    "        zeronet_monthly.iloc[i, -2] = zeronet_monthly.iloc[i-1, 3]-zeronet_monthly.iloc[i, 3]\n",
    "        \n",
    "for i in range(0, len(zeronet_monthly)):\n",
    "    if zeronet_monthly.iloc[i,9] == \"long\":\n",
    "        zeronet_monthly.iloc[i, -1] = zeronet_monthly.iloc[i, 4]-zeronet_monthly.iloc[i+1, 4]\n",
    "    else:\n",
    "        zeronet_monthly.iloc[i, -1] = zeronet_monthly.iloc[i-1, 4]-zeronet_monthly.iloc[i, 4]\n",
    "\n",
    "#Create two new columns containing predicted the exess return of the portfolio and initially set the values to zero.\n",
    "zeronet_monthly[\"pred_excess_return_zeronet_ew\"] =0\n",
    "zeronet_monthly[\"pred_excess_return_zeronet_vw\"] =0\n",
    "\n",
    "# predicted excess return zeronet in t = (weigted predicted excess return long in t) - (weigted predicted excess return short in t)\n",
    "for i in range(0, len(zeronet_monthly)):\n",
    "    if zeronet_monthly.iloc[i,9] == \"long\":\n",
    "        zeronet_monthly.iloc[i, -2] = zeronet_monthly.iloc[i, 5]-zeronet_monthly.iloc[i+1, 5]\n",
    "    else:\n",
    "        zeronet_monthly.iloc[i, -2] = zeronet_monthly.iloc[i-1, 5]-zeronet_monthly.iloc[i, 5]\n",
    "        \n",
    "for i in range(0, len(zeronet_monthly)):\n",
    "    if zeronet_monthly.iloc[i,9] == \"long\":\n",
    "        zeronet_monthly.iloc[i, -1] = zeronet_monthly.iloc[i, 6]-zeronet_monthly.iloc[i+1, 6]\n",
    "    else:\n",
    "        zeronet_monthly.iloc[i, -1] = zeronet_monthly.iloc[i-1, 6]-zeronet_monthly.iloc[i, 6]\n",
    "\n",
    "#Create two new columns containing return of the portfolio and initially set the values to zero.       \n",
    "zeronet_monthly[\"return_zeronet_ew\"] =0\n",
    "zeronet_monthly[\"return_zeronet_vw\"] =0\n",
    "\n",
    "# return zeronet in t = (weigted return long in t) - (weigted return short in t)\n",
    "for i in range(0, len(zeronet_monthly)):\n",
    "    if zeronet_monthly.iloc[i,9] == \"long\":\n",
    "        zeronet_monthly.iloc[i, -2] = zeronet_monthly.iloc[i, 7]-zeronet_monthly.iloc[i+1, 7]\n",
    "    else:\n",
    "        zeronet_monthly.iloc[i, -2] = zeronet_monthly.iloc[i-1, 7]-zeronet_monthly.iloc[i, 7]\n",
    "        \n",
    "for i in range(0, len(zeronet_monthly)):\n",
    "    if zeronet_monthly.iloc[i,9] == \"long\":\n",
    "        zeronet_monthly.iloc[i, -1] = zeronet_monthly.iloc[i, 8]-zeronet_monthly.iloc[i+1, 8]\n",
    "    else:\n",
    "        zeronet_monthly.iloc[i, -1] = zeronet_monthly.iloc[i-1, 8]-zeronet_monthly.iloc[i, 8]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NumMonth</th>\n",
       "      <th>MonthYear1</th>\n",
       "      <th>excess_return_zeronet_ew</th>\n",
       "      <th>excess_return_zeronet_vw</th>\n",
       "      <th>pred_excess_return_zeronet_ew</th>\n",
       "      <th>pred_excess_return_zeronet_vw</th>\n",
       "      <th>return_zeronet_ew</th>\n",
       "      <th>return_zeronet_vw</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-01</td>\n",
       "      <td>0.046050</td>\n",
       "      <td>0.015270</td>\n",
       "      <td>0.020668</td>\n",
       "      <td>0.019658</td>\n",
       "      <td>0.046351</td>\n",
       "      <td>0.015315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>747</th>\n",
       "      <td>2</td>\n",
       "      <td>2010-02</td>\n",
       "      <td>0.024549</td>\n",
       "      <td>0.007160</td>\n",
       "      <td>0.020912</td>\n",
       "      <td>0.019417</td>\n",
       "      <td>0.022035</td>\n",
       "      <td>0.007144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1477</th>\n",
       "      <td>3</td>\n",
       "      <td>2010-03</td>\n",
       "      <td>0.013240</td>\n",
       "      <td>-0.016004</td>\n",
       "      <td>0.020910</td>\n",
       "      <td>0.020297</td>\n",
       "      <td>0.014477</td>\n",
       "      <td>-0.015995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2227</th>\n",
       "      <td>4</td>\n",
       "      <td>2010-04</td>\n",
       "      <td>0.012271</td>\n",
       "      <td>0.021173</td>\n",
       "      <td>0.027081</td>\n",
       "      <td>0.025333</td>\n",
       "      <td>0.016102</td>\n",
       "      <td>0.022890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2992</th>\n",
       "      <td>5</td>\n",
       "      <td>2010-05</td>\n",
       "      <td>0.036214</td>\n",
       "      <td>0.000973</td>\n",
       "      <td>0.020526</td>\n",
       "      <td>0.019110</td>\n",
       "      <td>0.032085</td>\n",
       "      <td>0.000936</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89890</th>\n",
       "      <td>116</td>\n",
       "      <td>2019-08</td>\n",
       "      <td>-0.016163</td>\n",
       "      <td>-0.004743</td>\n",
       "      <td>0.010577</td>\n",
       "      <td>0.008493</td>\n",
       "      <td>-0.019153</td>\n",
       "      <td>-0.004775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90595</th>\n",
       "      <td>117</td>\n",
       "      <td>2019-09</td>\n",
       "      <td>0.036713</td>\n",
       "      <td>0.022036</td>\n",
       "      <td>0.010116</td>\n",
       "      <td>0.008668</td>\n",
       "      <td>0.036713</td>\n",
       "      <td>0.022036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91303</th>\n",
       "      <td>118</td>\n",
       "      <td>2019-10</td>\n",
       "      <td>0.015487</td>\n",
       "      <td>-0.002095</td>\n",
       "      <td>0.010482</td>\n",
       "      <td>0.008919</td>\n",
       "      <td>0.011009</td>\n",
       "      <td>-0.002185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91977</th>\n",
       "      <td>119</td>\n",
       "      <td>2019-11</td>\n",
       "      <td>0.047530</td>\n",
       "      <td>0.034357</td>\n",
       "      <td>0.012197</td>\n",
       "      <td>0.009831</td>\n",
       "      <td>0.049930</td>\n",
       "      <td>0.035645</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92694</th>\n",
       "      <td>120</td>\n",
       "      <td>2019-12</td>\n",
       "      <td>0.017637</td>\n",
       "      <td>0.029913</td>\n",
       "      <td>0.011543</td>\n",
       "      <td>0.010226</td>\n",
       "      <td>0.015677</td>\n",
       "      <td>0.029461</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>120 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       NumMonth MonthYear1  excess_return_zeronet_ew  \\\n",
       "16            1    2010-01                  0.046050   \n",
       "747           2    2010-02                  0.024549   \n",
       "1477          3    2010-03                  0.013240   \n",
       "2227          4    2010-04                  0.012271   \n",
       "2992          5    2010-05                  0.036214   \n",
       "...         ...        ...                       ...   \n",
       "89890       116    2019-08                 -0.016163   \n",
       "90595       117    2019-09                  0.036713   \n",
       "91303       118    2019-10                  0.015487   \n",
       "91977       119    2019-11                  0.047530   \n",
       "92694       120    2019-12                  0.017637   \n",
       "\n",
       "       excess_return_zeronet_vw  pred_excess_return_zeronet_ew  \\\n",
       "16                     0.015270                       0.020668   \n",
       "747                    0.007160                       0.020912   \n",
       "1477                  -0.016004                       0.020910   \n",
       "2227                   0.021173                       0.027081   \n",
       "2992                   0.000973                       0.020526   \n",
       "...                         ...                            ...   \n",
       "89890                 -0.004743                       0.010577   \n",
       "90595                  0.022036                       0.010116   \n",
       "91303                 -0.002095                       0.010482   \n",
       "91977                  0.034357                       0.012197   \n",
       "92694                  0.029913                       0.011543   \n",
       "\n",
       "       pred_excess_return_zeronet_vw  return_zeronet_ew  return_zeronet_vw  \n",
       "16                          0.019658           0.046351           0.015315  \n",
       "747                         0.019417           0.022035           0.007144  \n",
       "1477                        0.020297           0.014477          -0.015995  \n",
       "2227                        0.025333           0.016102           0.022890  \n",
       "2992                        0.019110           0.032085           0.000936  \n",
       "...                              ...                ...                ...  \n",
       "89890                       0.008493          -0.019153          -0.004775  \n",
       "90595                       0.008668           0.036713           0.022036  \n",
       "91303                       0.008919           0.011009          -0.002185  \n",
       "91977                       0.009831           0.049930           0.035645  \n",
       "92694                       0.010226           0.015677           0.029461  \n",
       "\n",
       "[120 rows x 8 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Only the measures at portfolio level are needed\n",
    "zeronet_monthly = zeronet_monthly[['NumMonth', 'MonthYear1', 'excess_return_zeronet_ew',\n",
    "                                   'excess_return_zeronet_vw', 'pred_excess_return_zeronet_ew',\n",
    "                                   'pred_excess_return_zeronet_vw','return_zeronet_ew',\n",
    "                                   'return_zeronet_vw']].drop_duplicates()\n",
    "\n",
    "zeronet_monthly"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Calculate zero-net portfolio performance measures                                            \n",
    "#Time-series average of realized excess returns                                             \n",
    "ew_mean_return_zeronet= zeronet_monthly[\"excess_return_zeronet_ew\"].mean()\n",
    "vw_mean_return_zeronet= zeronet_monthly[\"excess_return_zeronet_vw\"].mean()\n",
    "#Time-series average of predicted excess returns\n",
    "ew_mean_pred_return_zeronet = zeronet_monthly[\"pred_excess_return_zeronet_ew\"].mean()\n",
    "vw_mean_pred_return_zeronet = zeronet_monthly[\"pred_excess_return_zeronet_vw\"].mean()\n",
    "#Standard deviation of realized excess returns\n",
    "std_ew_zeronet = zeronet_monthly[\"excess_return_zeronet_ew\"].std()\n",
    "std_vw_zeronet = zeronet_monthly[\"excess_return_zeronet_vw\"].std()\n",
    "#Annualized sharpe ratio of realized excess returns\n",
    "sharpe_ew_zeronet = (zeronet_monthly[\"excess_return_zeronet_ew\"].mean()/zeronet_monthly[\"return_zeronet_ew\"].std())* np.sqrt(12)\n",
    "sharpe_vw_zeronet = (zeronet_monthly[\"excess_return_zeronet_vw\"].mean()/zeronet_monthly[\"return_zeronet_vw\"].std())* np.sqrt(12)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate table containing the performance measures for each decile portfolio and for the long-short portfolio for each weignting sheme respectively\n",
    "\n",
    "#### Equally weighted table:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pred</th>\n",
       "      <th>Real</th>\n",
       "      <th>Std</th>\n",
       "      <th>Sharpe</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Low (L)</th>\n",
       "      <td>-0.41%</td>\n",
       "      <td>-0.33%</td>\n",
       "      <td>4.75%</td>\n",
       "      <td>-0.23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.22%</td>\n",
       "      <td>0.21%</td>\n",
       "      <td>4.25%</td>\n",
       "      <td>0.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.46%</td>\n",
       "      <td>0.24%</td>\n",
       "      <td>4.42%</td>\n",
       "      <td>0.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.63%</td>\n",
       "      <td>0.56%</td>\n",
       "      <td>4.18%</td>\n",
       "      <td>0.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.78%</td>\n",
       "      <td>0.74%</td>\n",
       "      <td>4.34%</td>\n",
       "      <td>0.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.91%</td>\n",
       "      <td>0.74%</td>\n",
       "      <td>4.34%</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.04%</td>\n",
       "      <td>0.70%</td>\n",
       "      <td>4.47%</td>\n",
       "      <td>0.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.17%</td>\n",
       "      <td>1.07%</td>\n",
       "      <td>4.32%</td>\n",
       "      <td>0.85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.32%</td>\n",
       "      <td>1.10%</td>\n",
       "      <td>4.75%</td>\n",
       "      <td>0.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>High (H)</th>\n",
       "      <td>1.61%</td>\n",
       "      <td>1.18%</td>\n",
       "      <td>4.82%</td>\n",
       "      <td>0.84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>H-L</th>\n",
       "      <td>2.02%</td>\n",
       "      <td>1.50%</td>\n",
       "      <td>2.51%</td>\n",
       "      <td>2.01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Pred    Real    Std Sharpe\n",
       "Low (L)   -0.41%  -0.33%  4.75%  -0.23\n",
       "2          0.22%   0.21%  4.25%   0.17\n",
       "3          0.46%   0.24%  4.42%   0.19\n",
       "4          0.63%   0.56%  4.18%   0.46\n",
       "5          0.78%   0.74%  4.34%   0.58\n",
       "6          0.91%   0.74%  4.34%   0.59\n",
       "7          1.04%   0.70%  4.47%   0.54\n",
       "8          1.17%   1.07%  4.32%   0.85\n",
       "9          1.32%   1.10%  4.75%   0.79\n",
       "High (H)   1.61%   1.18%  4.82%   0.84\n",
       "H-L        2.02%   1.50%  2.51%   2.01"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chart_np = np.array([[ew_mean_pred_return_rank_0, ew_mean_return_rank_0, std_ew_rank_0, sharpe_ew_rank_0],\n",
    "                     [ew_mean_pred_return_rank_1, ew_mean_return_rank_1, std_ew_rank_1, sharpe_ew_rank_1],\n",
    "                     [ew_mean_pred_return_rank_2, ew_mean_return_rank_2, std_ew_rank_2, sharpe_ew_rank_2],\n",
    "                     [ew_mean_pred_return_rank_3, ew_mean_return_rank_3, std_ew_rank_3, sharpe_ew_rank_3],\n",
    "                     [ew_mean_pred_return_rank_4, ew_mean_return_rank_4, std_ew_rank_4, sharpe_ew_rank_4],\n",
    "                     [ew_mean_pred_return_rank_5, ew_mean_return_rank_5, std_ew_rank_5, sharpe_ew_rank_5],\n",
    "                     [ew_mean_pred_return_rank_6, ew_mean_return_rank_6, std_ew_rank_6, sharpe_ew_rank_6],\n",
    "                     [ew_mean_pred_return_rank_7, ew_mean_return_rank_7, std_ew_rank_7, sharpe_ew_rank_7],\n",
    "                     [ew_mean_pred_return_rank_8, ew_mean_return_rank_8, std_ew_rank_8, sharpe_ew_rank_8],\n",
    "                     [ew_mean_pred_return_rank_9, ew_mean_return_rank_9, std_ew_rank_9, sharpe_ew_rank_9],\n",
    "                     [ew_mean_pred_return_zeronet, ew_mean_return_zeronet, std_ew_zeronet, sharpe_ew_zeronet]])\n",
    "\n",
    "ew_df = pd.DataFrame(chart_np, columns=['Pred', 'Real', 'Std', 'Sharpe'],\n",
    "                              index=['Low (L)', '2', '3', '4', '5','6','7','8',\"9\",'High (H)', \"H-L\"])\n",
    "\n",
    "ew_df['Pred'] = pd.Series([\"{0:.2f}%\".format(val * 100) for val in ew_df['Pred']], index = ew_df.index)\n",
    "ew_df['Real'] = pd.Series([\"{0:.2f}%\".format(val * 100) for val in ew_df['Real']], index = ew_df.index)\n",
    "ew_df['Std'] = pd.Series([\"{0:.2f}%\".format(val * 100) for val in ew_df['Std']], index = ew_df.index)\n",
    "ew_df['Sharpe'] = pd.Series([(\"%.2f\" % round(val, 2)) for val in ew_df['Sharpe']], index = ew_df.index)\n",
    "ew_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Value weighted table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pred</th>\n",
       "      <th>Real</th>\n",
       "      <th>Std</th>\n",
       "      <th>Sharpe</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Low (L)</th>\n",
       "      <td>-0.36%</td>\n",
       "      <td>0.36%</td>\n",
       "      <td>4.75%</td>\n",
       "      <td>0.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.22%</td>\n",
       "      <td>0.38%</td>\n",
       "      <td>5.14%</td>\n",
       "      <td>0.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.46%</td>\n",
       "      <td>0.31%</td>\n",
       "      <td>5.53%</td>\n",
       "      <td>0.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.63%</td>\n",
       "      <td>0.66%</td>\n",
       "      <td>5.04%</td>\n",
       "      <td>0.45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.78%</td>\n",
       "      <td>0.60%</td>\n",
       "      <td>4.80%</td>\n",
       "      <td>0.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.91%</td>\n",
       "      <td>0.53%</td>\n",
       "      <td>4.99%</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.04%</td>\n",
       "      <td>0.38%</td>\n",
       "      <td>4.95%</td>\n",
       "      <td>0.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.16%</td>\n",
       "      <td>0.73%</td>\n",
       "      <td>4.52%</td>\n",
       "      <td>0.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.32%</td>\n",
       "      <td>0.63%</td>\n",
       "      <td>4.75%</td>\n",
       "      <td>0.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>High (H)</th>\n",
       "      <td>1.58%</td>\n",
       "      <td>0.73%</td>\n",
       "      <td>4.93%</td>\n",
       "      <td>0.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>H-L</th>\n",
       "      <td>1.94%</td>\n",
       "      <td>0.37%</td>\n",
       "      <td>3.07%</td>\n",
       "      <td>0.41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Pred   Real    Std Sharpe\n",
       "Low (L)   -0.36%  0.36%  4.75%   0.26\n",
       "2          0.22%  0.38%  5.14%   0.26\n",
       "3          0.46%  0.31%  5.53%   0.19\n",
       "4          0.63%  0.66%  5.04%   0.45\n",
       "5          0.78%  0.60%  4.80%   0.43\n",
       "6          0.91%  0.53%  4.99%   0.36\n",
       "7          1.04%  0.38%  4.95%   0.26\n",
       "8          1.16%  0.73%  4.52%   0.56\n",
       "9          1.32%  0.63%  4.75%   0.46\n",
       "High (H)   1.58%  0.73%  4.93%   0.50\n",
       "H-L        1.94%  0.37%  3.07%   0.41"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chart_np = np.array([[vw_mean_pred_return_rank_0, vw_mean_return_rank_0, std_vw_rank_0, sharpe_vw_rank_0],\n",
    "                     [vw_mean_pred_return_rank_1, vw_mean_return_rank_1, std_vw_rank_1, sharpe_vw_rank_1],\n",
    "                     [vw_mean_pred_return_rank_2, vw_mean_return_rank_2, std_vw_rank_2, sharpe_vw_rank_2],\n",
    "                     [vw_mean_pred_return_rank_3, vw_mean_return_rank_3, std_vw_rank_3, sharpe_vw_rank_3],\n",
    "                     [vw_mean_pred_return_rank_4, vw_mean_return_rank_4, std_vw_rank_4, sharpe_vw_rank_4],\n",
    "                     [vw_mean_pred_return_rank_5, vw_mean_return_rank_5, std_vw_rank_5, sharpe_vw_rank_5],\n",
    "                     [vw_mean_pred_return_rank_6, vw_mean_return_rank_6, std_vw_rank_6, sharpe_vw_rank_6],\n",
    "                     [vw_mean_pred_return_rank_7, vw_mean_return_rank_7, std_vw_rank_7, sharpe_vw_rank_7],\n",
    "                     [vw_mean_pred_return_rank_8, vw_mean_return_rank_8, std_vw_rank_8, sharpe_vw_rank_8],\n",
    "                     [vw_mean_pred_return_rank_9, vw_mean_return_rank_9, std_vw_rank_9, sharpe_vw_rank_9],\n",
    "                     [vw_mean_pred_return_zeronet, vw_mean_return_zeronet, std_vw_zeronet, sharpe_vw_zeronet]])\n",
    "\n",
    "vw_df = pd.DataFrame(chart_np, columns=['Pred', 'Real', 'Std', 'Sharpe'],\n",
    "                              index=['Low (L)', '2', '3', '4', '5','6','7','8',\"9\",'High (H)', \"H-L\"])\n",
    "\n",
    "vw_df['Pred'] = pd.Series([\"{0:.2f}%\".format(val * 100) for val in vw_df['Pred']], index = vw_df.index)\n",
    "vw_df['Real'] = pd.Series([\"{0:.2f}%\".format(val * 100) for val in vw_df['Real']], index = vw_df.index)\n",
    "vw_df['Std'] = pd.Series([\"{0:.2f}%\".format(val * 100) for val in vw_df['Std']], index = vw_df.index)\n",
    "vw_df['Sharpe'] = pd.Series([(\"%.2f\" % round(val, 2)) for val in vw_df['Sharpe']], index = vw_df.index)\n",
    "vw_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Save to compare with the other ML models\n",
    "ew_df.to_csv(r'C:\\Users\\krist\\Documents\\Data\\decile_portfolios_results\\rf_ew.csv')\n",
    "vw_df.to_csv(r'C:\\Users\\krist\\Documents\\Data\\decile_portfolios_results\\rf_vw.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Additional performance measures of zero net investment long-short  portfolio with monthly restructuring, 2010:01-2019:12"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Drawdown, turnover, and risk-adjusted performance of the long-short portfolio"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Trunover\n",
    "\n",
    "Portfolio turnover:\n",
    "\n",
    "### $$\n",
    "\\text { Turnover }=\\frac{1}{T} \\sum_{t=1}^{T}\\left(\\sum_{i}\\left|w_{i, t+1}-w_{i, t}\\right|\\right)\n",
    "$$\n",
    ", where $w_{i, t}$ denotes the weight of stock $i$ in the portfolio in month $t$.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Only stock weights are needed for the calculation of the portfolio turnover \n",
    "long = rank_9.copy()\n",
    "long=long[[\"MonthYear1\",'NumMonth','id', 'eq_weights', 'me_weights']]\n",
    "short = rank_0.copy()\n",
    "short=short[[\"MonthYear1\",'NumMonth','id', 'eq_weights','me_weights']]\n",
    "short[\"eq_weights\"]=short[\"eq_weights\"]*(-1)\n",
    "short[\"me_weights\"]=short[\"me_weights\"]*(-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Create lead weight variables $w_{i, t+1}$ for the equal- uand value-weighting scheme (eq_weights_lead1 and me_weights_lead1)\n",
    "\n",
    "* lead weight variable = 0, when the stock is not included in the portfolio in the subsequent month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Equal weights in t+1 for the top decile portfolio\n",
    "def create_lead_eq_weights(long):\n",
    "    mn = long.NumMonth.min()\n",
    "    mx = long.NumMonth.max() + 1\n",
    "    d = long.set_index('NumMonth').reindex(pd.RangeIndex(mn, mx, name='NumMonth'))\n",
    "    return pd.concat([d, d['eq_weights'].shift(-1).rename('eq_weights_lead1')], axis=1)\n",
    "\n",
    "eq_weights_lead1_long=long.groupby('id')[['NumMonth', 'eq_weights']].apply(create_lead_eq_weights) \\\n",
    "    .sort_index(level=[1, 0]).dropna(subset=['eq_weights']).reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Value weights in t+1 for the top decile portfolio\n",
    "def create_lead_me_weights(long):\n",
    "    mn = long.NumMonth.min()\n",
    "    mx = long.NumMonth.max() + 1\n",
    "    d = long.set_index('NumMonth').reindex(pd.RangeIndex(mn, mx, name='NumMonth'))\n",
    "    return pd.concat([d, d['me_weights'].shift(-1).rename('me_weights_lead1')], axis=1)\n",
    "\n",
    "me_weights_lead1_long=long.groupby('id')[['NumMonth', 'me_weights']].apply(create_lead_me_weights) \\\n",
    "    .sort_index(level=[1, 0]).dropna(subset=['me_weights']).reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Equal weights in t+1 for the bottom decile portfolio\n",
    "def create_lead_eq_weights(short):\n",
    "    mn = short.NumMonth.min()\n",
    "    mx = short.NumMonth.max() + 1\n",
    "    d = short.set_index('NumMonth').reindex(pd.RangeIndex(mn, mx, name='NumMonth'))\n",
    "    return pd.concat([d, d['eq_weights'].shift(-1).rename('eq_weights_lead1')], axis=1)\n",
    "\n",
    "eq_weights_lead1_short=short.groupby('id')[['NumMonth', 'eq_weights']].apply(create_lead_eq_weights) \\\n",
    "    .sort_index(level=[1, 0]).dropna(subset=['eq_weights']).reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Value weights in t+1 for the bottom decile portfolio\n",
    "def create_lead_me_weights(short):\n",
    "    mn = short.NumMonth.min()\n",
    "    mx = short.NumMonth.max() + 1\n",
    "    d = short.set_index('NumMonth').reindex(pd.RangeIndex(mn, mx, name='NumMonth'))\n",
    "    return pd.concat([d, d['me_weights'].shift(-1).rename('me_weights_lead1')], axis=1)\n",
    "\n",
    "me_weights_lead1_short=short.groupby('id')[['NumMonth', 'me_weights']].apply(create_lead_me_weights) \\\n",
    "    .sort_index(level=[1, 0]).dropna(subset=['me_weights']).reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Merge lead weight variables to the dataframe containing the top decile portfolio\n",
    "long=long.set_index(['NumMonth','id'])\n",
    "eq_weights_lead1_long=eq_weights_lead1_long.set_index(['NumMonth','id'])                    \n",
    "me_weights_lead1_long=me_weights_lead1_long.set_index(['NumMonth','id']) \n",
    "long = pd.merge(long, eq_weights_lead1_long[[\"eq_weights_lead1\"]],left_index=True, right_index=True)\n",
    "long = pd.merge(long, me_weights_lead1_long[[\"me_weights_lead1\"]],left_index=True, right_index=True)\n",
    "long=long.reset_index()\n",
    "\n",
    "# Merge lead weight variables to the dataframe containing the bottom decile portfolio\n",
    "short=short.set_index(['NumMonth','id'])\n",
    "eq_weights_lead1_short=eq_weights_lead1_short.set_index(['NumMonth','id'])                    \n",
    "me_weights_lead1_short=me_weights_lead1_short.set_index(['NumMonth','id']) \n",
    "short = pd.merge(short, eq_weights_lead1_short[[\"eq_weights_lead1\"]],left_index=True, right_index=True)\n",
    "short = pd.merge(short, me_weights_lead1_short[[\"me_weights_lead1\"]],left_index=True, right_index=True)\n",
    "short=short.reset_index()\n",
    "\n",
    "# Merge to get the zeronet portfolio\n",
    "long[\"Strategy\"]= \"long\"\n",
    "short[\"Strategy\"]= \"short\"\n",
    "zeronet= pd.concat([long, short])\n",
    "zeronet = zeronet.sort_values(by = ['NumMonth',\"id\",\"Strategy\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Absolute difference between the weight of the stocks in the two following months $\\left|w_{i, t+1}-w_{i, t}\\right|$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "zeronet[\"abs_ew\"] = abs(zeronet[\"eq_weights_lead1\"] -zeronet[\"eq_weights\"])\n",
    "zeronet[\"abs_vw\"] = abs(zeronet[\"me_weights_lead1\"] -zeronet[\"me_weights\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Sum of the absolute diffrernces over all stocks per month $\\sum_{i}\\left|w_{i, t+1}-w_{i, t}\\right|$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "zeronet[\"sum_abs_ew\"] = zeronet.groupby('NumMonth', sort=False)[\"abs_ew\"].transform('sum')\n",
    "zeronet[\"sum_abs_vw\"] = zeronet.groupby('NumMonth', sort=False)[\"abs_vw\"].transform('sum')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-  $\\text { Turnover }=\\frac{1}{T} \\sum_{t=1}^{T}\\left(\\sum_{i}\\left|w_{i, t+1}-w_{i, t}\\right|\\right)$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "turnover_ew = (zeronet[[\"NumMonth\",\"sum_abs_ew\"]].drop_duplicates().sum_abs_ew.sum())/zeronet[\"NumMonth\"].nunique()\n",
    "turnover_vw = (zeronet[[\"NumMonth\",\"sum_abs_vw\"]].drop_duplicates().sum_abs_vw.sum())/zeronet[\"NumMonth\"].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Multiply by 100 to get percent values\n",
    "turnover_ew_pct = turnover_ew*100\n",
    "turnover_vw_pct = turnover_vw*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trunover equally weighted:  1.7218053841789143\n",
      "Trunover value weighted:  22.22960631052653\n"
     ]
    }
   ],
   "source": [
    "print(\"Trunover equally weighted: \", turnover_ew_pct)\n",
    "print(\"Trunover value weighted: \", turnover_vw_pct)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Maximum Drawdown \n",
    "\n",
    "$$\n",
    "\\operatorname{MaxDD} \\max _{0 \\leq t_{1} \\leq t_{2} \\leq T}\\left(Y_{t_{1}}-Y_{t_{2}}\\right)\n",
    "$$\n",
    ", where $Y_{t}$ is the cumulative log return from date 0 through $t$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MaxDD equally weighted:  9.924583708620238\n",
      "MaxDD value weighted:  22.253459475156\n"
     ]
    }
   ],
   "source": [
    "def MDD(returns):\n",
    " \n",
    "    cum_rets = (1 + np.log(returns+1)).cumprod() - 1\n",
    "    nav = ((1 + cum_rets) * 100).fillna(100)\n",
    "    hwm = nav.cummax()\n",
    "    dd = nav / hwm - 1\n",
    "\n",
    "    return min(dd)\n",
    "\n",
    "maxDD_ew_pct = (MDD(zeronet_monthly[\"excess_return_zeronet_ew\"])*-100)\n",
    "maxDD_vw_pct = (MDD(zeronet_monthly[\"excess_return_zeronet_vw\"])*-100)    \n",
    "\n",
    "print(\"MaxDD equally weighted: \", maxDD_ew_pct)\n",
    "print(\"MaxDD value weighted: \", maxDD_vw_pct)       \n",
    "           "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Max 1M loss (%)\n",
    "\n",
    "- the most extreme negative monthly return "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Max 1M loss equally weighted:  3.859410020835107\n",
      "Max 1M loss value weighted:  7.438617710304767\n"
     ]
    }
   ],
   "source": [
    "max_1m_loss_ew= (zeronet_monthly[\"excess_return_zeronet_ew\"].min())*(-100)\n",
    "max_1m_loss_vw= (zeronet_monthly[\"excess_return_zeronet_vw\"].min())*(-100)\n",
    "\n",
    "print(\"Max 1M loss equally weighted: \", max_1m_loss_ew)\n",
    "print(\"Max 1M loss value weighted: \", max_1m_loss_vw)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Risk-adjusted performance\n",
    "\n",
    "Mean active return and information ratio relative to DAX (benchmark)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "#get DAX price data\n",
    "dax = pd.read_csv(r\"C:\\Users\\krist\\Documents\\Data\\Dax monthly.csv\",parse_dates =[\"Date\"])\n",
    "dax= dax[dax['Date'] <= '2020-02-01']\n",
    "\n",
    "# Share prices are given on the first day of each month and in the data set at the end of the month. \n",
    "# Shift by one month to match the months in both datasets \n",
    "dax['MonthYear_t+1'] = dax['Date'].dt.to_period('M')\n",
    "dax['MonthYear'] = dax[\"MonthYear_t+1\"].shift(1)\n",
    "dax.drop([\"MonthYear_t+1\"], axis = 1, inplace = True)\n",
    "\n",
    "# We only need the prices from January 2010\n",
    "dax= dax[dax['MonthYear'] >= '2010-01']\n",
    "\n",
    "# Set MonthYear as index to  merge with the portfolio dataframe afterwards\n",
    "dax=dax.set_index([\"MonthYear\"])\n",
    "\n",
    "# Get monthly returns\n",
    "dax[\"monthly_return_dax\"] = dax[\"Close\"].pct_change()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Get Dax excess returns\n",
    "risk_free = bigdata[[\"MonthYear1\", \"risk_free_rate\"]].drop_duplicates()\n",
    "risk_free=risk_free.set_index(\"MonthYear1\")\n",
    "dax = pd.merge(dax[[\"monthly_return_dax\"]], risk_free ,left_index=True, right_index=True)\n",
    "dax[\"monthly_excess_return_dax\"] = dax[\"monthly_return_dax\"]- dax[\"risk_free_rate\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Merge zeronet and DAX dataframes\n",
    "zeronet_monthly=zeronet_monthly.set_index(\"MonthYear1\")\n",
    "dax_zeronet = pd.merge(dax[[\"monthly_return_dax\",\"monthly_excess_return_dax\"]], zeronet_monthly[[\"excess_return_zeronet_ew\", \"excess_return_zeronet_vw\", \"return_zeronet_ew\", \"return_zeronet_vw\"]],left_index=True, right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate portfolio performance measures\n",
    "\n",
    "#Time-series average of portfolio excess_returns \n",
    "mean_ret_ew = dax_zeronet[\"excess_return_zeronet_ew\"].mean()\n",
    "mean_ret_vw = dax_zeronet[\"excess_return_zeronet_vw\"].mean()\n",
    "mean_ret_dax = dax_zeronet[\"monthly_excess_return_dax\"].mean()\n",
    "#Multiply by 100 to get percent values\n",
    "mean_ret_ew_pct = mean_ret_ew*100\n",
    "mean_ret_vw_pct= mean_ret_vw*100\n",
    "mean_ret_dax_pct = mean_ret_dax*100\n",
    "\n",
    "#Standard deviation ofportfolio excess_returns\n",
    "std_ew = dax_zeronet[\"excess_return_zeronet_ew\"].std()\n",
    "std_vw = dax_zeronet[\"excess_return_zeronet_vw\"].std()\n",
    "std_dax = dax_zeronet[\"monthly_excess_return_dax\"].std()\n",
    "#Multiply by 100 to get percent values\n",
    "std_ew_pct = std_ew*100\n",
    "std_vw_pct = std_vw*100\n",
    "std_dax_pct = std_dax*100\n",
    "\n",
    "#Annualized sharpe ratio of portfolio excess_returns\n",
    "sr_ew = ((dax_zeronet[\"excess_return_zeronet_ew\"]).mean() / (dax_zeronet[\"return_zeronet_ew\"]).std()) * np.sqrt(12)\n",
    "sr_vw = ((dax_zeronet[\"excess_return_zeronet_vw\"]).mean() / (dax_zeronet[\"return_zeronet_vw\"]).std()) * np.sqrt(12)\n",
    "sr_dax = ((dax_zeronet[\"monthly_excess_return_dax\"]).mean() / (dax_zeronet[\"monthly_return_dax\"]).std()) * np.sqrt(12)\n",
    "#Multiply by 100 to get percent values\n",
    "sr_ew_pct = sr_ew*100\n",
    "sr_vw_pct = sr_vw*100\n",
    "sr_dax_pcr= sr_dax*100\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mean active return:  difference between the benchmark return and the portfolio return"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_act_ret_ew = (dax_zeronet[\"excess_return_zeronet_ew\"]-dax_zeronet[\"monthly_excess_return_dax\"]).mean()\n",
    "mean_act_ret_vw = (dax_zeronet[\"excess_return_zeronet_vw\"]-dax_zeronet[\"monthly_excess_return_dax\"]).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## Information Ratio $=\\frac{R_{p}-R_{b}}{\\sigma_{\\left(R_{p}-R_{b}\\right)}}$\n",
    "- $R_{p}-$ Return of Portfolio\n",
    "- $R_{b}-$ Return of Benchmark (Dax)\n",
    "- $\\sigma_{\\left(R_{p}-R_{b}\\right)}-$ Tracking Error\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "ir_ew = ((dax_zeronet[\"excess_return_zeronet_ew\"]-dax_zeronet[\"monthly_excess_return_dax\"]).mean() / \n",
    "         (dax_zeronet[\"excess_return_zeronet_ew\"]-dax_zeronet[\"monthly_excess_return_dax\"]).std())\n",
    "\n",
    "ir_vw = ((dax_zeronet[\"excess_return_zeronet_vw\"]-dax_zeronet[\"monthly_excess_return_dax\"]).mean() / \n",
    "         (dax_zeronet[\"excess_return_zeronet_vw\"]-dax_zeronet[\"monthly_excess_return_dax\"]).std()) \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "# This measures we only need to represent for the long-short portfolio \n",
    "turnover_dax=0\n",
    "ir_dax=0\n",
    "mean_act_ret_dax=0\n",
    "maxDD_dax=0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>DAX</th>\n",
       "      <th>RF</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Mean return [%]</th>\n",
       "      <td>0.752467</td>\n",
       "      <td>1.504461</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Std [%]</th>\n",
       "      <td>4.610969</td>\n",
       "      <td>2.507019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sharpe ratio</th>\n",
       "      <td>0.561110</td>\n",
       "      <td>2.008944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maximum drawdown [%]</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.924584</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Turnover [%]</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.721805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mean active return</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.007259</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Information ratio</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.146268</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           DAX        RF\n",
       "Mean return [%]       0.752467  1.504461\n",
       "Std [%]               4.610969  2.507019\n",
       "Sharpe ratio          0.561110  2.008944\n",
       "Maximum drawdown [%]  0.000000  9.924584\n",
       "Turnover [%]          0.000000  1.721805\n",
       "Mean active return    0.000000  0.007259\n",
       "Information ratio     0.000000  0.146268"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generate table representing the portfolio performance measures for the equal weighting scheme\n",
    "ew_chart = np.array([[mean_ret_dax_pct, mean_ret_ew_pct],\n",
    "                     [std_dax_pct,std_ew_pct],\n",
    "                     [sr_dax, sr_ew],\n",
    "                     [maxDD_dax, maxDD_ew_pct],\n",
    "                     [turnover_dax, turnover_ew_pct],\n",
    "                     [mean_act_ret_dax, mean_act_ret_ew],\n",
    "                     [ir_dax, ir_ew]])\n",
    "\n",
    "\n",
    "ew_chart = pd.DataFrame(ew_chart, columns=['DAX', 'RF'],\n",
    "                        index=[\"Mean return [%]\",'Std [%]',\n",
    "                               \"Sharpe ratio\", \"Maximum drawdown [%]\",\"Turnover [%]\",\n",
    "                               \"Mean active return\",\"Information ratio\"])\n",
    "ew_chart"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>DAX</th>\n",
       "      <th>RF</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Mean return [%]</th>\n",
       "      <td>0.752467</td>\n",
       "      <td>0.366442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Std [%]</th>\n",
       "      <td>4.610969</td>\n",
       "      <td>3.067012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sharpe ratio</th>\n",
       "      <td>0.561110</td>\n",
       "      <td>0.405363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maximum drawdown [%]</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>22.253459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Turnover [%]</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>22.229606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mean active return</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.003958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Information ratio</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.073437</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           DAX         RF\n",
       "Mean return [%]       0.752467   0.366442\n",
       "Std [%]               4.610969   3.067012\n",
       "Sharpe ratio          0.561110   0.405363\n",
       "Maximum drawdown [%]  0.000000  22.253459\n",
       "Turnover [%]          0.000000  22.229606\n",
       "Mean active return    0.000000  -0.003958\n",
       "Information ratio     0.000000  -0.073437"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generate table representing the portfolio performance measures for the value weighting scheme\n",
    "vw_chart = np.array([[mean_ret_dax_pct, mean_ret_vw_pct],\n",
    "                     [std_dax_pct,std_vw_pct],\n",
    "                     [sr_dax, sr_vw],\n",
    "                     [maxDD_dax, maxDD_vw_pct],\n",
    "                     [turnover_dax, turnover_vw_pct],\n",
    "                     [mean_act_ret_dax, mean_act_ret_vw],\n",
    "                     [ir_dax, ir_vw]])\n",
    "\n",
    "\n",
    "vw_chart = pd.DataFrame(vw_chart, columns=['DAX', 'RF'],\n",
    "                        index=[\"Mean return [%]\",'Std [%]',\n",
    "                               \"Sharpe ratio\", \"Maximum drawdown [%]\",\"Turnover [%]\",\n",
    "                               \"Mean active return\",\"Information ratio\"])\n",
    "vw_chart"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save results to compare with the other ML models\n",
    "ew_chart.to_csv(r'C:\\Users\\krist\\Documents\\Data\\performance_zeronet\\rf_zeronet_ew.csv')\n",
    "vw_chart.to_csv(r'C:\\Users\\krist\\Documents\\Data\\performance_zeronet\\rf_zeronet_vw.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Cumulative performance of long-only investment portfolio with monthly restructuring, 2010:01-2019:12\n",
    "\n",
    "The H-L returns,  tend to be driven more by the long side (top decile) than the short side (bottom decile), so the long-only portfolio performace is also examined separately."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Turnover"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trunover equally weighted:  0.8430195536472564\n",
      "Trunover value weighted:  10.927435557047525\n"
     ]
    }
   ],
   "source": [
    "long[\"abs_ew\"] = abs(long[\"eq_weights_lead1\"] -long[\"eq_weights\"])\n",
    "long[\"abs_vw\"] = abs(long[\"me_weights_lead1\"] -long[\"me_weights\"])\n",
    "long[\"sum_abs_ew\"] = long.groupby('NumMonth', sort=False)[\"abs_ew\"].transform('sum')\n",
    "long[\"sum_abs_vw\"] = long.groupby('NumMonth', sort=False)[\"abs_vw\"].transform('sum')\n",
    "turnover_ew_long = (long[[\"NumMonth\",\"sum_abs_ew\"]].drop_duplicates().sum_abs_ew.sum())/long[\"NumMonth\"].nunique()\n",
    "turnover_vw_long = (long[[\"NumMonth\",\"sum_abs_vw\"]].drop_duplicates().sum_abs_vw.sum())/long[\"NumMonth\"].nunique()\n",
    "turnover_ew_pct_long = turnover_ew_long*100\n",
    "turnover_vw_pct_long = turnover_vw_long*100\n",
    "print(\"Trunover equally weighted: \", turnover_ew_pct_long)\n",
    "print(\"Trunover value weighted: \", turnover_vw_pct_long)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Maximum Drawdown"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MaxDD equally weighted:  31.43310768415839\n",
      "MaxDD value weighted:  30.709102456849745\n"
     ]
    }
   ],
   "source": [
    "long_monthly = rank_9[['MonthYear1','NumMonth', 'return_portfolio_ew', 'return_portfolio_vw', \"excess_return_portfolio_ew\", \"excess_return_portfolio_vw\"]].drop_duplicates()\n",
    "\n",
    "maxDD_ew_pct_long = (MDD(long_monthly[\"excess_return_portfolio_ew\"])*-100)\n",
    "maxDD_vw_pct_long = (MDD(long_monthly[\"excess_return_portfolio_vw\"])*-100)    \n",
    "\n",
    "print(\"MaxDD equally weighted: \", maxDD_ew_pct_long)\n",
    "print(\"MaxDD value weighted: \", maxDD_vw_pct_long)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Max 1M loss (%)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Max 1M loss equally weighted:  12.494533768371687\n",
      "Max 1M loss value weighted:  15.872361078675427\n"
     ]
    }
   ],
   "source": [
    "max_1m_loss_ew_long= (long_monthly[\"excess_return_portfolio_ew\"].min())*(-100)\n",
    "max_1m_loss_vw_long= (long_monthly[\"excess_return_portfolio_vw\"].min())*(-100)\n",
    "\n",
    "print(\"Max 1M loss equally weighted: \", max_1m_loss_ew_long)\n",
    "print(\"Max 1M loss value weighted: \", max_1m_loss_vw_long)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "long_monthly=long_monthly.set_index(\"MonthYear1\")\n",
    "dax_long = pd.merge(dax[[\"monthly_return_dax\",\"monthly_excess_return_dax\"]], long_monthly[[\"return_portfolio_ew\", \"return_portfolio_vw\", \"excess_return_portfolio_ew\", \"excess_return_portfolio_vw\"]],left_index=True, right_index=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Time-series average of portfolio excess returns \n",
    "mean_ret_ew_long = dax_long[\"excess_return_portfolio_ew\"].mean()\n",
    "mean_ret_vw_long = dax_long[\"excess_return_portfolio_vw\"].mean()\n",
    "mean_ret_dax = dax_long[\"monthly_excess_return_dax\"].mean()\n",
    "#Multiply by 100 to get percent values\n",
    "mean_ret_ew_long_pct = mean_ret_ew_long*100\n",
    "mean_ret_vw_long_pct= mean_ret_vw_long*100\n",
    "mean_ret_dax_pct = mean_ret_dax*100\n",
    "\n",
    "#Standard deviation ofportfolio excess returns\n",
    "std_ew_long = dax_long[\"excess_return_portfolio_ew\"].std()\n",
    "std_vw_long = dax_long[\"excess_return_portfolio_vw\"].std()\n",
    "std_dax = dax_long[\"monthly_excess_return_dax\"].std()\n",
    "#Multiply by 100 to get percent values\n",
    "std_ew_long_pct = std_ew_long*100\n",
    "std_vw_long_pct = std_vw_long*100\n",
    "std_dax_pct = std_dax*100\n",
    "\n",
    "#Annualized sharpe ratio of portfolio excess returns\n",
    "sr_ew_long = ((dax_long[\"excess_return_portfolio_ew\"]).mean() / (dax_long[\"return_portfolio_ew\"]).std()) * np.sqrt(12)\n",
    "sr_vw_long = ((dax_long[\"excess_return_portfolio_vw\"]).mean() / (dax_long[\"return_portfolio_vw\"]).std()) * np.sqrt(12)\n",
    "sr_dax = ((dax_long[\"monthly_excess_return_dax\"]).mean() / (dax_long[\"monthly_return_dax\"]).std()) * np.sqrt(12)\n",
    "#Multiply by 100 to get percent values\n",
    "sr_ew_long_pct = sr_ew_long*100\n",
    "sr_vw_long_pct = sr_vw_long*100\n",
    "sr_dax_pcr= sr_dax*100\n",
    "\n",
    "#Mean active excess_return:  difference between the benchmark excess return and the portfolio excess return\n",
    "mean_act_ret_ew_long = (dax_long[\"excess_return_portfolio_ew\"]-dax_long[\"monthly_excess_return_dax\"]).mean()\n",
    "mean_act_ret_vw_long = (dax_long[\"excess_return_portfolio_vw\"]-dax_long[\"monthly_excess_return_dax\"]).mean()\n",
    "\n",
    "# Calculate the information ratio\n",
    "ir_ew_long = ((dax_long[\"excess_return_portfolio_ew\"]-dax_long[\"monthly_excess_return_dax\"]).mean() / \n",
    "         (dax_long[\"excess_return_portfolio_ew\"]-dax_long[\"monthly_excess_return_dax\"]).std()) \n",
    "\n",
    "ir_vw_long = ((dax_long[\"excess_return_portfolio_vw\"]-dax_long[\"monthly_excess_return_dax\"]).mean() / \n",
    "         (dax_long[\"excess_return_portfolio_vw\"]-dax_long[\"monthly_excess_return_dax\"]).std()) \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>DAX</th>\n",
       "      <th>RF_long</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Mean return [%]</th>\n",
       "      <td>0.752467</td>\n",
       "      <td>1.178145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Std [%]</th>\n",
       "      <td>4.610969</td>\n",
       "      <td>4.821312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sharpe ratio</th>\n",
       "      <td>0.561110</td>\n",
       "      <td>0.836973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maximum drawdown [%]</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.433108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Turnover [%]</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.843020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mean active return</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.004274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Information ratio</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.120300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           DAX    RF_long\n",
       "Mean return [%]       0.752467   1.178145\n",
       "Std [%]               4.610969   4.821312\n",
       "Sharpe ratio          0.561110   0.836973\n",
       "Maximum drawdown [%]  0.000000  31.433108\n",
       "Turnover [%]          0.000000   0.843020\n",
       "Mean active return    0.000000   0.004274\n",
       "Information ratio     0.000000   0.120300"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generate table representing the portfolio performance measures for the equal weighting scheme\n",
    "ew_long_chart = np.array([[mean_ret_dax_pct, mean_ret_ew_long_pct],\n",
    "                     [std_dax_pct,std_ew_long_pct],\n",
    "                     [sr_dax, sr_ew_long],\n",
    "                     [maxDD_dax, maxDD_ew_pct_long],\n",
    "                     [turnover_dax, turnover_ew_pct_long],\n",
    "                     [mean_act_ret_dax, mean_act_ret_ew_long],\n",
    "                     [ir_dax, ir_ew_long]])\n",
    "\n",
    "\n",
    "ew_long_chart = pd.DataFrame(ew_long_chart, columns=['DAX', 'RF_long'],\n",
    "                        index=[\"Mean return [%]\",'Std [%]',\n",
    "                               \"Sharpe ratio\", \"Maximum drawdown [%]\",\"Turnover [%]\",\n",
    "                               \"Mean active return\",\"Information ratio\"])\n",
    "\n",
    "ew_long_chart"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>DAX</th>\n",
       "      <th>RF_long</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Mean return [%]</th>\n",
       "      <td>0.752467</td>\n",
       "      <td>0.725484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Std [%]</th>\n",
       "      <td>4.610969</td>\n",
       "      <td>4.929937</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sharpe ratio</th>\n",
       "      <td>0.561110</td>\n",
       "      <td>0.504099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maximum drawdown [%]</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>30.709102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Turnover [%]</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>10.927436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mean active return</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Information ratio</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000404</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           DAX    RF_long\n",
       "Mean return [%]       0.752467   0.725484\n",
       "Std [%]               4.610969   4.929937\n",
       "Sharpe ratio          0.561110   0.504099\n",
       "Maximum drawdown [%]  0.000000  30.709102\n",
       "Turnover [%]          0.000000  10.927436\n",
       "Mean active return    0.000000   0.000015\n",
       "Information ratio     0.000000   0.000404"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generate table representing the portfolio performance measures for the value weighting scheme\n",
    "vw_long_chart = np.array([[mean_ret_dax_pct, mean_ret_vw_long_pct],\n",
    "                     [std_dax_pct,std_vw_long_pct],\n",
    "                     [sr_dax, sr_vw_long],\n",
    "                     [maxDD_dax, maxDD_vw_pct_long],\n",
    "                     [turnover_dax, turnover_vw_pct_long],\n",
    "                     [mean_act_ret_dax, mean_act_ret_vw_long],\n",
    "                     [ir_dax, ir_vw_long]])\n",
    "\n",
    "\n",
    "vw_long_chart = pd.DataFrame(vw_long_chart, columns=['DAX', 'RF_long'],\n",
    "                        index=[\"Mean return [%]\",'Std [%]',\n",
    "                               \"Sharpe ratio\", \"Maximum drawdown [%]\",\"Turnover [%]\",\n",
    "                               \"Mean active return\",\"Information ratio\"])\n",
    "\n",
    "vw_long_chart"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save results to compare with the other ML models\n",
    "ew_long_chart.to_csv(r'C:\\Users\\krist\\Documents\\Data\\performance_long_only\\rf_long_ew.csv')\n",
    "vw_long_chart.to_csv(r'C:\\Users\\krist\\Documents\\Data\\performance_long_only\\rf_long_vw.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Cumulative performance graph of expected return-sorted portfolios with monthly restructuring, 2010:01-2019:12\n",
    "\n",
    "\n",
    "Calculate the cumulative log returns of the long (top decile) and short (bottom decile) portfolios for both weighting schemes. Let's assume one invests 1 USD in January 2010 and buys a fraction of the top decile portfolio.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "long =rank_9.copy()\n",
    "long = long.set_index(\"MonthYear1\")\n",
    "long=long[[\"excess_return_portfolio_ew\", \"excess_return_portfolio_vw\"]].drop_duplicates()\n",
    "\n",
    "#Shift to get the returns in the current month\n",
    "long[\"curr_ret_port_ew\"]=long['excess_return_portfolio_ew'].shift(1)\n",
    "long[\"curr_ret_port_vw\"]=long['excess_return_portfolio_vw'].shift(1)\n",
    "\n",
    "#Calculate cumulative returns and cumulative log returns for the equal weighting scheme\n",
    "long[\"cumulative_return_ew\"]= (1 + long[\"curr_ret_port_ew\"]).cumprod()-1\n",
    "long[\"cumulative_log_return_ew\"]= (1 + np.log(long[\"curr_ret_port_ew\"]+1)).cumprod()-1\n",
    "# return in January 2010 = 0 (one invests in the end of the month)\n",
    "long.iloc[0, 2:] = 0\n",
    "#Plot cumulative log returns for the equal weighting scheme\n",
    "long[[\"cumulative_log_return_ew\"]].plot(grid=True,figsize=(15, 7))\n",
    "\n",
    "#Calculate cumulative returns and cumulative log returns for the value weighting scheme\n",
    "long[\"cumulative_return_vw\"]= (1 + long[\"curr_ret_port_vw\"]).cumprod()-1\n",
    "long[\"cumulative_log_return_vw\"]= (1 + np.log(long[\"curr_ret_port_vw\"]+1)).cumprod()-1\n",
    "# Return in January 2010 = 0(one invests in the end of the month)\n",
    "long.iloc[0, 2:] = 0\n",
    "# Plot cumulative log returns for the  value weighting scheme\n",
    "long[[\"cumulative_log_return_vw\"]].plot(grid=True,figsize=(15, 7));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "short =rank_0.copy()\n",
    "short = short.set_index(\"MonthYear1\")\n",
    "short=short[[\"excess_return_portfolio_ew\", \"excess_return_portfolio_vw\"]].drop_duplicates()\n",
    "\n",
    "#Shift to get the returns in the current month\n",
    "short[\"curr_ret_port_ew\"]=short['excess_return_portfolio_ew'].shift(1)\n",
    "short[\"curr_ret_port_vw\"]=short['excess_return_portfolio_vw'].shift(1)\n",
    "\n",
    "#Calculate cumulative returns and cumulative log returns for the equal weighting scheme\n",
    "short[\"cumulative_return_ew\"]= (1 + short[\"curr_ret_port_ew\"]).cumprod()-1\n",
    "short[\"cumulative_log_return_ew\"]= (1 + np.log(short[\"curr_ret_port_ew\"]+1)).cumprod()-1\n",
    "# return in January 2010 = 0 (one invests in the end of the month)\n",
    "short.iloc[0, 2:] = 0\n",
    "#Plot cumulative log returns for the equal weighting scheme\n",
    "short[[\"cumulative_log_return_ew\"]].plot(grid=True,figsize=(15, 7))\n",
    "\n",
    "#Calculate cumulative returns and cumulative log returns for the value weighting scheme\n",
    "short[\"cumulative_return_vw\"]= (1 + short[\"curr_ret_port_vw\"]).cumprod()-1\n",
    "short[\"cumulative_log_return_vw\"]= (1 + np.log(short[\"curr_ret_port_vw\"]+1)).cumprod()-1\n",
    "# Return in January 2010 = 0(one invests in the end of the month)\n",
    "short.iloc[0, 2:] = 0\n",
    "# Plot cumulative log returns for the  value weighting scheme\n",
    "short[[\"cumulative_log_return_vw\"]].plot(grid=True,figsize=(15, 7));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Merge results to compare cumulative returns of both portfolios for the equal weighting scheme\n",
    "long = long.rename(columns={'cumulative_log_return_ew': 'cumulative_log_return_long_ew'})\n",
    "short = short.rename(columns={'cumulative_log_return_ew': 'cumulative_log_return_short_ew'})\n",
    "long_short_ew = pd.merge(long[[\"cumulative_log_return_long_ew\"]],short[[\"cumulative_log_return_short_ew\"]],left_index=True, right_index=True)\n",
    "#Plot\n",
    "long_short_ew[[\"cumulative_log_return_long_ew\", \"cumulative_log_return_short_ew\"]].plot(grid=True,figsize=(17, 5));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Merge results to compare cumulative returns of both portfolios for the value weighting scheme\n",
    "long = long.rename(columns={'cumulative_log_return_vw': 'cumulative_log_return_long_vw'})\n",
    "short = short.rename(columns={'cumulative_log_return_vw': 'cumulative_log_return_short_vw'})\n",
    "long_short_vw = pd.merge(long[[\"cumulative_log_return_long_vw\"]],short[[\"cumulative_log_return_short_vw\"]],left_index=True, right_index=True)\n",
    "#Plot\n",
    "long_short_vw[[\"cumulative_log_return_long_vw\", \"cumulative_log_return_short_vw\"]].plot(grid=True,figsize=(17, 5));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DAX Returns\n",
    "\n",
    "Compare the cumulative returns of both portfolios with DAX cumulative returs over the same period"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# calculate cumulative returns\n",
    "dax[\"cumulative_return\"]= (1 + dax[\"monthly_excess_return_dax\"]).cumprod()-1\n",
    "dax[\"cumulative_log_return\"]= (1 + np.log(dax[\"monthly_excess_return_dax\"]+1)).cumprod()-1\n",
    "\n",
    "#replace NaN with 0: return and cumulative return on Jan 2010=0\n",
    "dax.iloc[0,4:]=0\n",
    "#plot\n",
    "dax[[\"cumulative_log_return\"]].plot(grid=True,figsize=(15, 7));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3IAAAIZCAYAAADjkw3MAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzdd3hc1bXw4d8Uzaj33nuxLVuW5d6wjQvgUEzvwdTQEr6EkJtAAgGSGyDJhSQQQi+hmo5tjHHHXbIkq1jN6r33OjPn+8PYYKyukUZlvc/Dgz3nnH3W2ZKsWbP3XlulKIqCEEIIIYQQQogJQ23pAIQQQgghhBBCDI0kckIIIYQQQggxwUgiJ4QQQgghhBATjCRyQgghhBBCCDHBSCInhBBCCCGEEBOMJHJCCCGEEEIIMcFIIieEEEIIIYQQE4zW0gH0p6GhDZNJtrkzFzc3e+rqWi0dxqQh/Wl+0qfmJf1pXtKf5id9al7Sn+YnfWpe0p9Do1arcHGx6/P4uE7kTCZFEjkzk/40L+lP85M+NS/pT/OS/jQ/6VPzkv40P+lT85L+NB+ZWimEEEIIIYQQE4wkckIIIYQQQggxwUgiJ4QQQgghhBATzLheI9cbo9FAQ0MNBkO3pUOZcKqr1ZhMJkuHMWmMp/7UanW4uHig0Uy4H2khhBBCCDEME+5dX0NDDdbWttjZeaNSqSwdzoSi1aoxGMZH4jEZjJf+VBSFtrZmGhpqcHf3sXQ4QgghhBBiDEy4qZUGQzd2do6SxAnxHZVKhZ2do4xSCyGEEEJMIRMukQMkiRPiR+RnQgghhBBiapmQidxUdOxYIvfee0e/5+zfv4/33nsbgE8/3cSnn24a0/uPlszMdJ5//rlRvYcln08IIYQQQoihmnBr5ETfsrIyz/z50kuvsGAk5lVYWEBDQ72lwxBCCCGEEGLckERuhBRF4YUX/sHevbvRajVcfPEG9u7dxcaNdxAfn0BFRTn33XcnmzZ9wZNPPoq1tQ05OVm0trZwxx33sG3bFvLycli69Dzuu+8Btmz5guTkJH73u0cBuPfeO9i48eyRouTkJP7zn+fp6uqkpaWV++9/AH//QD777GMAvL19qKysAMDR0YnS0mIeeODXAPzjH3/H09OTn/zkMv72t7+Qn38Sk8nE9dffxOrV6wb1zMXFRTz11JO0tDRjbW3DL37xK2JiplNdXcUf//gILS3NhIaGk5JyjE8+2dJnO1u2fMHWrV/S1NTI4sXLuPLKa3j66T9RVVWFWq3mzjvvITp6Gi+//G86Ojp4441X8PDw7LN/XnjhOYxGE6GhYfj4+FJbW0NJSTFVVZWsX38JN99864ie78knH8XOzp7s7BPU1tZw6623s27dT2htbeWJJ35PaWkpvr5+1NRU8ac/PYOPj2+f93jrrdfZtWs7RqOJ+fMX8LOf3c9DD/0/LrvsChYuXMyLL/6LnJxs/vrX56itreWBB+7mrbc+GFT8QgghhBBi8pvQidz+tAq+PV4xKm0vmenD4tiBKwDu2rWDtLRU3nzzPQwGA3fffRvd3V19nl9bW8OLL77G1q1f8uc/P8a7736MXq/n0ksv5JZbbh9UbB999D6/+c0jBAUFk5R0lGeffYY333yfSy7ZAMBFF13MK6+8CMD5569l48bruf/+X6LRqNizZycvvvgab7zxClFRMTz88GO0tbVy110bmTZtBn5+/gPe//HHH+GGG37K8uUrSU9P4+GHH+Lddz/m2WefYeXK1WzYcCV79uxi+/avBmyrpqaat9/+EK1Wyx/+8D9cdNHFLFmynNraWu6++1Zef/0dbrvtLpKTk7j55lvZsuWLPtsqKSlm06Yvsbe355VXXiQvL5fnn3+Z1tYWrrrqUjZsuAoHB4dhPx9AdXUVzz//Mvn5J7nvvjtZt+4nvPbaSwQGBvG///s3srIyufPOW/pt/9ChA2Rnn+Cll95EpVLx+OO/5+uvt7Jo0WKSko6ycOFiUlOTqa6uwmg0cvjwARYuXDJg3EIIIYQQYuqY0InceJCSksTKlavR6XTodDpef/2dftdaLViwCAAvL29CQsJwcXEFwNHRkZaW5kHd85FHHufAgX3s2vUNGRlpdHR09Hmui4sL4eERHDuWiLW1jsDAINzc3ElMPEJXVyebN38OQGdnJwUF+QMmcu3t7ZSWlrJ8+UoAZsyIxdHRkeLiIo4ePXJmpGz58hXY2w+cNEVGRqPVnvo2TEw8QlFRES+/fCoJNRgMlJWVDtjGaQEBQdjb25/5e3x8AlZWVri4uOLo6EhbW+uAiVx/zwcwb958VCoVoaFhNDc3fRf3YX7/+ycAiI6eRmhoWL/3SEw8QmZmOrfeeiMAXV2deHl5c9llV/Cb3/w/2tvbAAgPjyAnJ4tDhw5wxRVXD7ofhBBCCCHE5DehE7nFsYMbNRtNWq2WHxYMrKgoP6uCoMFgOOt8KyurM3/WaDTntKdSqVAU5czfjUbDOefcc8/txMfPYfbsOcyZM5fHHnu43xjXrr2QnTu3o9NZsWbNBQCYTEYeeeRxoqKiAaivr8PR0anfdgAU5dx90xQFjEYjarUak0np5aq+6fX6M382Gk0899wLZ+Kora3FxcWF3NycM+f01z8/bAtAp9P1eV1f+nu+U23qz7R32qnnHvx+ciaTkauuupZrrrkBgJaWFjQaDba2tphMCrt37yQ2dhaurq4kJh4lO/sEM2bMHHT7QgghhBBi8pOqlSM0a1Y8u3fvxGAw0NnZyS9/eR/29g4UFJwEYN++3UNqz8nJmaKiAhRFoby8jLy8vLOONzc3UVJSxK233sWCBYvZt2/PmSRCo9GcSTh+aOnS5aSkHOPw4UMsW7YCgPj4uWeqWtbW1nLzzddSVVU5YHx2dvb4+vqxZ89OANLT06ivryM0NIyEhHlnplMePLif1taWIT37nDkJfPzxhwAUFORz001X09XVedZzDdQ/I9Xf8/UlIWH+mec+eTKP/PyT/W4HEB8/l23bttDe3o7BYOB//ueX7N69Azg1YvvGG68we/Yc4uPn8tFH7zN9emyvSb8QQgghhJi6JvSI3HiwfPkKsrIy2bjxekwmhSuvvJbo6BiefPJRNm/+nKVLzxtSewkJ89i8+TOuvfZygoKCmDkz7qzjjo5OrF9/CTfeeBVarZb4+Ll0dnbS0dFBXFw8Tz75KK6urmddo9dbExs7C4PBgK2tLQAbN97OX//6F2688SpMJhN3333/oNbHAfz+94/z9NN/4pVXXsTKSseTTz6FlZUVv/jFr3j88T/w+ecfEx4eOaiplT/0wAO/5qmnnuTmm69BURQeeeSP2NraERMznVdf/Q8vvPAPbr31zn77xxz6er6+/PSnt/KnPz3GzTdfg6+vP25u7ueMDv7QkiXLyMvL4Y47forJZGT+/EVccMF6ABYtWsJ7773NzJlx2NjYYDD0sGiRrI8TQgghhBBnUymDmW9mIXV1redM1ausLMLbO8hCEU1sWq0ag2HwUwCH6sMP3yMhYR4hIaFkZ2fxl788wauvvj1q97O00/25bdsWfHx8mTkzjsrKSu677w7ef/9T1OqxHfCeDD8bHh4O1NQMbSRX9E3607ykP81P+tS8pD/NT/rUvKQ/h0atVuHmZt/ncRmRE2ekpibz978/3euxZ555Fnd3j36v9/cP4NFHf4darUKn0/PQQw+zY8fXvPXW672e//rr74w05CEZ6fP1JSgomKef/jMmkxGVSs2DD/6WtLTUUbmXEEIIIYQQICNyU8poj8hNNeOtPyfDz4Z8Umde0p/mJf1pftKn5iX9aX7Sp+Yl/Tk0A43ISbETIYQQQgghfiQ5t4a2jh5LhyFEnySRE0IIIYQQ4geqGzv4x0dpfLb3pKVDEaJPksgJIYQQQgjxAwXlzQAcz6u1cCRC9E0SOSGEEEIIIX6goOJUIpddVE9X97l79AoxHkgiJ4QQQgghxA8UVjSj1agxGBVySxstHY4QvZJEboI4diyRe++9o99z9u/fx3vvndq37dNPN/Hpp5vG9P6jJTMzneeff84i977iip9QUVE+7OtfeeVFUlOTzRiREEIIIUaTyaRQVNXKguleaDUqThQ1WDokIXol+8hNIllZmWf+fOmlV1gwEvMqLCygoaHe0mEMS3JyErNnz7F0GEIIIYQYpPK6Nrp6jEQFONPQ2i2JnBi3JJEbIUVReOGFf7B37260Wg0XX7yBvXt3sXHjHcTHJ1BRUc59993Jpk1f8OSTj2JtbUNOThatrS3cccc9bNu2hby8HJYuPY/77nuALVu+IDk5id/97lEA7r33DjZuPHskLDk5if/853m6ujppaWnl/vsfwN8/kM8++xgAb28fKisrAHB0dKK0tJgHHvg1AP/4x9/x9PTkJz+5jL/97S/k55/EZDJx/fU3sXr1ukE9c3FxEU899SQtLc1YW9vwi1/8ipiY6VRXV/HHPz5CS0szoaHhpKQc45NPtvTZzpYtX7B165c0NTWyePEyrrzyGp5++k9UVVWhVqu58857iI6exssv/5uOjg7eeOMVPDw8++yfF154DqPRRGhoGD4+vtTW1lBSUkxVVSXr11/CzTff2mcsp2Pv6OhArVbx858/yIwZsQC89tpL5OZm09nZycMP/5Hp02dQXFzE00//iebmprP64MknH6WpqYmyshJuuOGnZGef4C9/eYI//ekZwsLCe713fX3dOc8dHh7JLbdcx6efbgXg0ksv4L77HmDVqjW89dZraDQarrvupkF9vYQQQggxeIUVp/Y5C/FxpL3HxHvbs2nr7MHO2srCkQlxtgmdyPXk7Kcne++otG0VtQyryMUDnrdr1w7S0lJ58833MBgM3H33bXR3d/V5fm1tDS+++Bpbt37Jn//8GO+++zF6vZ5LL72QW265fVCxffTR+/zmN48QFBRMUtJRnn32Gd58830uuWQDABdddDGvvPIiAOefv5aNG6/n/vt/iUajYs+enbz44mu88cYrREXF8PDDj9HW1spdd21k2rQZ+Pn5D3j/xx9/hBtu+CnLl68kPT2Nhx9+iHff/Zhnn32GlStXs2HDlezZs4vt278asK2ammrefvtDtFotf/jD/3DRRRezZMlyamtrufvuW3n99Xe47ba7SE5O4uabb2XLli/6bKukpJhNm77E3t6eV155kby8XJ5//mVaW1u46qpL2bDhKhwcHHq99ssvP2PRoiVcd91NHDp0gOPHU84kcsHBofz2t3/go4/e59133+KJJ/7C448/wk033cLSpSvO6gMAJycnnnrq7wBs3vw5Gzfe0WcSB/Dss8/0+tyenl7k5+eh0WgxGo0kJx9j1ao1HD58kF/96n8G7FshhBBCDF1BZTPWOg3ebrZodFre/TqbnOJGZkd6WDo0Ic4yoRO58SAlJYmVK1ej0+nQ6XS8/vo7/a4lW7BgEQBeXt6EhITh4uIKgKOjIy0tzYO65yOPPM6BA/vYtesbMjLS6Ojo6PNcFxcXwsMjOHYsEWtrHYGBQbi5uZOYeISurk42b/4cgM7OTgoK8gdM5Nrb2yktLWX58pUAzJgRi6OjI8XFRRw9euTMSNny5Suwt+89afqhyMhotNpT34aJiUcoKiri5ZdPJaEGg4GystIB2zgtICAIe3v7M3+Pj0/AysoKFxdXHB0daWtr7TORS0iYx+9+92tycrJZtGgJl19+1Zljy5adB0BISBi7d+880wcrVqzCYDCd1QcA06bNGHTM/T33woWLSUw8ilar5corr+Gbb7bR2tpKfX0dwcEhQ7qHEEIIIQansKKZYG8H1CoVUUEu6LRqThQ1SCInxp0JnchZRS4e1KjZaNJqtahU3/+9oqIc1Q9eMBgMZ51vZfX9sLxGozmnPZVKhaIoZ/5uNBrOOeeee24nPn4Os2fPYc6cuTz22MP9xrh27YXs3Lkdnc6KNWsuAMBkMvLII48TFRUNnJre5+jo1G87AIpi6uU1MBqNqNVqTCall6v6ptfrz/zZaDTx3HMvnImjtrYWFxcXcnNzzpzTX//8sC0AnU7X53U/NnNmHG+//QEHDnzLjh1fs2XLF/zf/z0PnP11UhSl3z7oLY6B9PXcRqOJV1/9Dzqdjttv/xk7d37D9u1fMW/ewiG1L4QQQojBMRhNlFS3cn5CAABWWg0R/k6cKJZ1cmL8kaqVIzRrVjy7d+/EYDDQ2dnJL395H/b2DhQUnARg377dQ2rPycmZoqICFEWhvLyMvLy8s443NzdRUlLErbfexYIFi9m3bw8m06nEQqPRnEkmfmjp0uWkpBzj8OFDLFu2AoD4+LlnqlrW1tZy883XUlVVOWB8dnb2+Pr6sWfPTgDS09Oor68jNDSMhIR5Z6ZTHjy4n9bWliE9+5w5CXz88YcAFBTkc9NNV9PV1XnWcw3UP8P1/PPPsm3bVi64YD0PPPAQOTnZfZ57ug927doBnN0HP3Z6WmR/+nruqKhoSkqKKCkpJigomPj4ObzxxissXrxkBE8qhBBCiL6UVLdiMCqE+DieeS06yIWymjaa2rotGJkQ55rQI3LjwfLlK8jKymTjxusxmRSuvPJaoqNjePLJR9m8+XOWLj1vSO0lJMxj8+bPuPbaywkKCmLmzLizjjs6OrF+/SXceONVaLVa4uPn0tnZSUdHB3Fx8Tz55KO4urqedY1eb01s7CwMBgO2trYAbNx4O3/961+48carMJlM3H33/YNaHwfw+98/ztNP/4lXXnkRKysdTz75FFZWVvziF7/i8cf/wOeff0x4eOSgplb+0AMP/JqnnnqSm2++BkVReOSRP2Jra0dMzHReffU/vPDCP7j11jv77Z/huvzyq3nssYfZsuUL1Go1Dz/8WL/n//73j/PMM3/mpZf+fVYf/Nj8+Qt55pk/8/DDjxEbO6vXtvp6bjg1UtjZeWrqbHz8XL744jPi4qQKphBCCDEaCr/bCDzE+/v3MDFBrkA+2cUNzIvxslBkQpxLpfQ338zC6upaz5mqV1lZhLd3kIUimti0WjUGw7nTAs3lww/fIyFhHiEhoWRnZ/GXvzzBq6++PWr3s7TR7s+hmgw/Gx4eDtTUDG0kV/RN+tO8pD/NT/rUvKQ/R+7VzSdIyavl2fuXoFKp8PBwoLKqifuf/ZZ5MZ7cvC7a0iFOaPI9OjRqtQo3N/s+j8uInDgjNTWZv//96V6PPfPMs7i797/I198/gEcf/R1qtQqdTs9DDz3Mjh1f89Zbr/d6/uuvvzPSkIdkpM83Ev/617McPXr4nNejo2P4zW8eGbX7CiGEEGLwCiqbCfZxOKvegUatJirAWfaTE+OOJHLijFmzZo8ouVq4cDELF55dfCYqKppVq9aMNDSzGOnzjcQ99/zcIvcVQgghxOB0dRspr20jPuLcD3ZjglxIyaulrqkTNydrC0QnxLmk2IkQQgghhJjyiqpaUBTOKnRyWkyQCwBZUr1SjCOSyAkhhBBCiCmv4HShE59zi7X5etjhYGsl0yvFuCKJnBBCCCGEmPIKKppxcdDjZH/ufrBqlYroQBdOFDX0uy+tEGNJEjkhhBBCCDHlFVa09Dqt8rSYIBcaWrqoaugYw6jEVHZ6lLgvkshNEMeOJXLvvXf0e87+/ft4771T5f4//XTTmQ2/x+r+oyUzM53nn39uTO/55JOPsmXLF2N6TyGEEEJYRmtHD9WNHb1Oqzzt9Do5mV4pxsqJwv6/16Rq5SSSlZV55s+XXnqFBSMxr8LCAhoa6i0dhhBCCCEmqcLKUyMfwf2MyHm62ODioOdEUQMrZvuNVWhiChtoRE4SuRFSFIUXXvgHe/fuRqvVcPHFG9i7dxcbN95BfHwCFRXl3HffnWza9AVPPvko1tY25ORk0drawh133MO2bVvIy8th6dLzuO++B9iy5QuSk5P43e8eBeDee+9g48azR8KSk5P4z3+ep6urk5aWVu6//wH8/QP57LOPAfD29qGysgIAR0cnSkuLeeCBXwPwj3/8HU9PT37yk8v429/+Qn7+SUwmE9dffxOrV68b1DMXFxfx1FNP0tLSjLW1Db/4xa+IiZlOdXUVf/zjI7S0NBMaGk5KyjE++WRLn+1s2fIFW7d+SVNTI4sXL+PKK6/h6af/RFVVFWq1mjvvvIfo6Gm8/PK/6ejo4I03XsHDw7PP/nnhhecwGk2Ehobh4+NLbW0NJSXFVFVVsn79Jdx88639fh3/+c+/s3//t7i7u2MymZg9ew4AL774L5KSjtLc3Iy7uzt//OOfqamp4cEHf84bb7yHRqPmlluu53//969EREQNqg+FEEIIMX4UVJzapDrEu+8ROZVKRUyQC8dP1mFSFNQ/2GtOCHPr6jFSWtPa7zkTOpE7XJHEwYqjo9L2Qp+5zPeZM+B5u3btIC0tlTfffA+DwcDdd99Gd3dXn+fX1tbw4ouvsXXrl/z5z4/x7rsfo9frufTSC7nlltsHFdtHH73Pb37zCEFBwSQlHeXZZ5/hzTff55JLNgBw0UUX88orLwJw/vlr2bjxeu6//5doNCr27NnJiy++xhtvvEJUVAwPP/wYbW2t3HXXRqZNm4Gfn/+A93/88Ue44Yafsnz5StLT03j44Yd4992PefbZZ1i5cjUbNlzJnj272L79qwHbqqmp5u23P0Sr1fKHP/wPF110MUuWLKe2tpa7776V119/h9tuu4vk5CRuvvnWfqc7lpQUs2nTl9jb2/PKKy+Sl5fL88+/TGtrC1dddSkbNlyFg0Pv/0Dv3r2DnJxs3n77A1paWvjpT68BoLS0hOLiQv7971dRq9U8/vjv2bZtK9deewOXXbaB559/FoPBwKWXXiFJnBBCCDFBFVY04+Vig621Vb/nxQS5cCC9krKaNgI87ccoOjEV5Zc1YTT1X1hnQidy40FKShIrV65Gp9Oh0+l4/fV3+l1LtmDBIgC8vLwJCQnDxcUVAEdHR1pa+h8+Pe2RRx7nwIF97Nr1DRkZaXR09L3o1sXFhfDwCI4dS8TaWkdgYBBubu4kJh6hq6uTzZs/B6Czs5OCgvwBE7n29nZKS0tZvnwlADNmxOLo6EhxcRFHjx45M1K2fPkK7O37/lTrtMjIaLTaU9+GiYlHKCoq4uWXTyWhBoOBsrLSAds4LSAgCHv77/9RjY9PwMrKChcXVxwdHWlra+0zkUtOTmL58hVotVpcXFxYsODUxub+/gHce+8DfPHFpxQXF5GRkXamj2655TZuvvkG9Ho9jzzyx0HHKYQQQojxpaCimehAlwHPO33OiaIGSeTEqMouaWSgQd8JncjN95kzqFGz0aTVas/q5IqKclQ/eMFgMJx1vpXV95/0aDSac9pTqVRnlbU1Gg3nnHPPPbcTHz+H2bPnMGfOXB577OF+Y1y79kJ27tyOTmfFmjUXAGAyGXnkkceJiooGoL6+DkdHp37bAVAUUy+vgdFoRK1WYxrgk4Mf0+u/L/FrNJp47rkXzsRRW1uLi4sLubk5Z87pr39+2BaATqfr87ofO3X8+7+f/tpkZZ3g0Ud/xzXXXMeKFavQaNRn2mltbaW9vZ329naam5txdnYewpMLIYQQYjxoaOmisbW73/Vxp7k5WePpYkNWUQNr5gaMQXRiqsotbcLHza7fc6Rq5QjNmhXP7t07MRgMdHZ28stf3oe9vQMFBScB2Ldv95Dac3JypqioAEVRKC8vIy8v76zjzc1NlJQUceutd7FgwWL27duDyXQqudJoNBiNxnPaXLp0OSkpxzh8+BDLlq0AID5+7pmqlrW1tdx887VUVVUOGJ+dnT2+vn7s2bMTgPT0NOrr6wgNDSMhYd6Z6ZQHD+6ntbVlSM8+Z04CH3/8IQAFBfncdNPVdHV1nvVcA/XPcCUkzGPnzu10d3fT3NzM4cMHgVMjrrNnz+HSS68gICCQAwe+PdPfTz/9v1x++ZVcdtkV/PWv/2uWOIQQQggxtgr72Qi8NzFBLmSXNGA0nfvhthDmYDCaOFnWRIhv/x8uTOgRufFg+fIVZGVlsnHj9ZhMCldeeS3R0TE8+eSjbN78OUuXnjek9hIS5rF582dce+3lBAUFMXNm3FnHHR2dWL/+Em688Sq0Wi3x8XPp7Oyko6ODuLh4nnzyUVxdXc+6Rq+3JjZ2FgaDAVtbWwA2brydv/71L9x441WYTCbuvvv+Qa2PA/j97x/n6af/xCuvvIiVlY4nn3wKKysrfvGLX/H443/g888/Jjw8clBTK3/ogQd+zVNPPcnNN1+Doig88sgfsbW1IyZmOq+++h9eeOEf3Hrrnf32z3AtXXoeJ05kctNNV+Pq6kZwcCgAq1at4be/fZCbbroagKioGCoqytmxYzulpSX8/vdPoCgKt912Ezt2bGfVqtVmiUcIIYQQY6OgsgW1SkWg1+ATuT0p5RRVthI6wBttIYajsLKFboOJUO/+v79Uygi2p//nP//J1q1bAVi+fDm//vWvzzp+4sQJfve739HW1kZCQgKPPfbYmfVQg1FX13rOVL3KyiK8vYOGG/KUptWqMRhG79OjDz98j4SEeYSEhJKdncVf/vIEr7769qjdz9JGuz+HajL8bHh4OFBTM7SRXNE36U/zkv40P+lT85L+HJ6/vZ9CY2s3f7x13jnHeuvT5rZuHvjHt1y0KIgNy8LGKsxJQb5HB2froSI+3H2Sfz6wjCD/vtduDntE7sCBA3z77bd88sknqFQqbrvtNrZv387q1d+PSDz44IM88cQTxMXF8dvf/pYPPviA6667bri3FKMsNTWZv//96V6PPfPMs7i7e/R7vb9/AI8++jvUahU6nZ6HHnqYHTu+5q23Xu/1/Ndff2ekIQ/JSJ9PCCGEEJOLoigUVDQTHzn49wCOdjriItzZnVzO+oXB6KzOrXkgxEhklzTi42aLvU3/VVSHnch5eHjwm9/85kxBibCwMMrLy88cLysro7Ozk7i4OAA2bNjAc889J4ncODZr1uwRJVcLFy5m4cLFZ70WFRXNqlVrRhqaWYz0+YQQQggxudQ0ddLWaSBkEIVOfmjN3ACSc2s5kFHJeXGyObgwH5NJIbe0ibnRngOeO+xiJxEREWeStMLCQrZu3cry5cvPHK+ursbD4/tPNzw8PKiqqhru7YQQQgghhDCr7wudDC2RiwxwJsjbge1HSzANf5WSEOcorWmlo8tAVIDzgOeOuNhJbm4ud955J7/+9a8JDg4+87rJZDqrDL+iKGf9fTDc3M7dn6O6Wo1GoxpyW+IUrVYKlZrTeOlPRVFQq9V4eAytwMx4NBmeYTyR/jQv6U/zkz41L+nPoak8WIRWo2ZWjDdWffxO76tPr1gVyV//m0RJXQcJMV6jGeakIt+j/TuUVQPAgjg/3Fxs+z13RIlcUlIS999/P7/97W+56KKLzjrm7e1NTU3Nmb/X1tbi6TnwEOEP9VbsBFR0dXWj1fY/Z1Sca7wV55joxlN/Ggw9gGrCLyCWRdDmJf1pXtKf5id9al7Sn0OXmV9HgKc9jQ1tvR7vr0+jfB1wcdDzwfZsgtz7f8MtTpHv0YEdO1GJm6MelcFIXV1rrwNbpw17OKGiooJ77rmHZ5555pwkDsDPzw+9Xk9SUhIAn332GcuWLRvu7c6wsbGnpaWx142phZiKFMVES0sDNjZ9/6ALIYQQ4mwmk0JRVcug94/7Ma1Gzflz/DlR1EBxlSQnYuQURSGntInIQUyrhBGMyL3yyit0dXXxv//7/UbI11xzDTt37uT+++8nNjaWZ555hocffpjW1lamT5/OTTfdNNzbnWFv70RDQw1VVaWAzEkeCrVafWYzazFy46c/Veh01tjbO1k6ECGEEGLC+CaxhK5uI9GBfZd3H8iyOF8+31/I9qMl3Lp+mhmjE1NRVUMHzW3do5/IPfzwwzz88MPnvH7ttdee+XN0dDSbNm0a7i16pVKpcHUd2hRNcYoMZ5uX9KcQQggxMeWXN/Ph7pPMjnBnTtTwtx+ys7ZiyUwfdieXcfl5YTjb680YpZhqckoaAQadyI2PSg1CCCGEEEKMgfbOHv79WTrO9jpuuTBmxAX0Vif4YzIp7DxWaqYIxVSVU9KIg60V3q6DW3MpiZwQQgghhJgSFEXhta1ZNLR0ceclMwbccHkwPF1smR3pwa5jZXT1GM0QpZiqckoaiQxwHvSHC5LICSGEEEKIKWFXchlJ2TVsWB5KuJ/51pavmRtAW6eBA+mVZmtTTC31zZ3UNnUS6e886GskkRNCCCGEEJNecVUL7+3IJTbUjbXzAs3adoS/EyE+DnwtG4SLYRrq+jiQRE4IIYQQQkxyHV0GXvg0HXsbK25bH4N6hOvifkylUrFmbiBV9e0cz6sza9tiasgpacRGryHAc/DbSUkiJ4QQQgghJi1FUXhzWzbVjR3cefF0HGx1o3KfOVEeuDrq+fpo8ai0Lya3nNImwv2cUasH/yGDJHJCCCGEEGLS2ne8gsOZVVy6JISoEewZN5BTG4QHkFXcSF5Z06jdR0w+ze3dlNe2ERkwtHWbksgJIYQQQohJx6Qo7Egq5b/bc5gW7MJFC4NH/Z7LZvlgb2PFU+8c470dubR29Iz6PcXEl1tyKvEfyvo4GMGG4EIIIYQQQoxH5bVtvL41i7yyJqYHu3D7xdOHNGVtuGytrXhs4zw+2ZfP9qMl7E+r4KKFwaya44+VVsZPRO9ySxux0qoJ9nYc0nWSyAkhhBBCiEnBYDSx5VARXx4oRG+l4daLYlg0w3vEm34PhYuDno0XxrA6IYAPd+Xxwa48dh4r5fLlYcyL8RzTWMTEkF3SSJiv45CTfUnkhBBCCCHEhJdf3sxrW09QVtPGvBhPrj0/Eie70SlsMhgBnvb8v6vjSC+o44OdJ3nx8wy+PlrMuvlBzAxzQ2+lsVhsYvzo6DJQXNXC+mFM/ZVETgghhBBCTAgmk0JbZw+tHT/4r72HwsoWdqeU4Wyv5/7LZxIX4W7pUM+YEeLGtFtcOZBeySf78nnh03T0VhpmhbsxL8aL2FBXrLSS1E1VOSWNKApEBjoP+VpJ5IQQQgghxLhWUdfGsx8ep6axg96221YB583244rzwrDRj7+3t2q1iiUzfVg0w5vskkaOnqgiMbuGIyeqsdZpmB3hztzvkjqNWtbSTSX70yqws9YS6T+0ipUgiZwQQgghhBjn3t+ZR0tHNz9ZHIydjRUONlbY21hhb2uFvbUVDrY69LrxP6qlVquICXIhJsiF69dEcqKogSMnqknOqeFgRhUr4v24cU2UpcMUY6SptYvk3NrviuEM/ftXEjkhhBBCCDFupeXXcfxkHVetCGfd/EBLh2M2GrWaGSFuzAhxw7A2ire2ZbM3pZy18wLxdLaxdHhiDHybVoHRpLA8zndY18vYrRBCCCGEGJeMJhPv7cjF08WG8xP8LR3OqNFq1Fy6NBSVSsUX+wsGfZ2iKHy05ySHM6tGMToxGkyKwp6UcqIDnfFxsxtWG5LICSGEEEKIcWl3cjkVde1cvSIcrWZyv211cdCzYrYfB9IrqapvH9Q1Sdk1bD5YxGtbTlDT2DHKEQpzyiiop7apk/Nm+w27jcn9EyGEEEIIISakts4ePt2XT0yQy7iqQjmaLlwYhJVGzWeDGJXr6DLwzjc5+LrboVKreGtbNorSWykYMR7tTi7DwdaK+EiPYbchiZwQQgghhBh3Pv+2kPYuA9esipgym2g72elYOcefwxlVlNe29Xvup/sKaGrt5pYLo9mwNJT0gnqOZlWPUaRiJBpaukjNq2NJrM+IRpolkRNCCCGEEONKZX07O4+VsnSmLwGe9pYOZ0xdMD8QnU7DZ9/2PSpXVNnCN0klLJ/tR5ivE6vm+BPk7cA73+TS1tkzhtGK4dh3vByTMvwiJ6dJIieEEEIIIcaVD3bmYaVVc9myUEuHMuYcbHWcP8efo1nVlFa3nnPcZFJ4c1s2DjZWXL78VP+o1Sp+ui6alvZuPtp9cqxDFkNgMinsTS1nerALni62I2pLEjkhhBBCCDFuZBTUk5JXy08WBeNkp7N0OBaxdl4gNvreR+X2pJZTUNHM1SsjsLO2OvN6kLcDqxMC2J1STl5p01iGK4bgeH4d9c1dLI8bfpGT0ySRE0IIIYQQ44LRZOK9nbl4OFtzfkKApcOxGHsbK1YnBJCUU0NRZcuZ15vaTo24RQc6s2C61znXXbo0BFdHPW9sy8JgNI1lyGKQ9iSX4WSnM0sBH0nkhBBCCCHEuLA3tYKymjauWhGOlXZqv01dMzcQW732rFG5D3bm0tVj5Ma1Ub0WgLHWablhdRRlNW1sO1I8luGKQahr6uR4fh1LZ42syMlpU/snRAghhBBCjAv1zZ18sjefqADnEZVknyxsrbWsnR9ISl4t+eXNnCis52BGFRcsCOp3A+m4CHfmRHrw+f5CqhsGtx+dGBt7U8tBgWWzRlbk5DRJ5IQQQgghhEWl5tXy6GtH6TGYuPb8qbPdwEDOn+OPvY0VH+05yVtf5+DhbM36hUEDXnfd6kg0ahVvfZ0je8uNE0aTiX3Hy4kNc8PdycYsbUoiJ4QQQgghLMJgNPHejlye3XQcFwc9v/9pAoFeDpYOa9yw0Wu5YH4gJ4oaqKxv54Y1UeisNANe5+KgZ8OyUDIK6jmUUTUGkYqBpObV0djaPeItB35Ia7aWhBBCCCGEGKTqhnb+/VkGhZUtrIr356qVYVhpB05SppqV8f7sPFZGRIATsaFuQ7ru8IkqXv8qCzcnayIDnEcvSDGg3clluDjomRk2+K/hQGRETgghhBBCjKkjJ6p49LWjVDd0cM9lsVy/JlKSuD7odRqeuH0+t6+fNqTr1GoV910+EzdHa57dlHpW9UsxtmoaO8goqGfZLF80avOlX5LICSGEEEKIMWEwmnh9axb//iwDfw97Ht04lzlRUthkIHorzbDWDTra6vjVNXHY6LX87YMUKuul+IklHMqoBGDpTB+ztiuJnBBCCCGEGBOHMqrYm1rOBfMD+fV1s81W9EH0zdXRml9eHYeiwF/fS6a+udPSIU05idk1hPs74epobdZ2JZETQgghhBBj4mBGJZ7ONlxxXphZ9tESg+PjZscvr46jrdPAX99PoaW929IhTRlV9e2UVLcyJ8rT7G3LT5AQQgghhBh1DS1dZBU1sGC6l2wvYAFB3g78/IqZ1DR28vcPUunoMlg6pCkhMbsagDmjsDeiJHJCCCGEEGLUHTlRhQIsmO5t6VCmrKhAF+6+dAbFVa388+M0egxGS4c06SVm1xDq64ibk3mnVYIkckIIIYQQYgwcyqgi2NsBb1dbS4cypcVFuHPrRTGcKGrgw10nLR2ORRiMJuqaRn+tYG1jB0WVLSSMwrRKkH3khBBCCCHEKCuvbaOoqoVrVkVYOhQBLJzhzcGMSrJLGi0dikVsOVjEp98WEObnyHlxfsyN9hzURutDlZhdAzBqlVllRE4IIYQQQoyqQ5mVqFQwP2Z0RibE0Pm42VFV345JUSwdyphLPVmLu5M1bR0GXtl8gl/+az/v7ciloq7NrPdJyq4myMsBD+fRqc4qI3JCCCGEEGLUKIrCoYwqpgW54GSvt3Q44js+brZ0G0w0NHeNyvqt8aq1o4fCihYuXhLCxYuDyS5uZHdKGTuSSvn6aAnRgc6smx/IzDD3Ed2nvrmTk+XNXL481EyRn0tG5IQQQgghxKg5WdZMbVOnFDkZZ06vVZxqm4RnFTWgANOCXVCpVEQHuXDXJTN45p7FXL48lNqmTp798DjVDSPrl6Qz0ypHbxRaEjkhhBBCCDFqDmZWotOqiR+F8uti+LzdpmYil1lYj7VOQ4iP41mvO9npuGhhMP9zwxxQwbdpFSO6T2J2Nf4e9qNa3EcSOSGEEEIIMSoMRhNHT1QTF+GOjV5W9IwnTnY6rHUaKuumWiLXQHSgS58b0rs46IkNdWN/WiUm0/DWDza0dJFX2kRC9Oh+eCGJnBBCCCGEGBUZBfW0dvSwYJpMqxxvVCoV3q62VNSbt8DHeFbT2EF1YwfTgl36PW9JrA8NLV2kF9QN6z7HcmpQGN1plSCJnBBCCCGEGCWHMquws9YyI9TV0qGIXni72U6pqZWZhfUATAvu//sxLsIdB1sr9h0f3vTKpOxqfNxs8XO3G9b1gyWJnBBCCCGEGJLObgO7jpXS1W3s85yOLgPJOTXMjfHqcxqbsCxvV1vqm7v6/TpOJpmFDTjb6/Bx63/dmlajZuF0b1Jya2lu7x7SPZrbuskuaRy1TcB/SH6qhBBCCCHEkOxLreCtr3P483+TqG/u7PWc5Nwaug0mFkzzGuPoxGCdLsRRNcIKjROBSVE4UdTA9GBXVCrVgOcvneWL0aRwML1ySPc5lluDokBCtCRyQgghhBBinMkpacTOWkt1QwePv5FIfnnzOeccyqjCzdGacH8nC0QoBmMqbUFQUtVKa0fPgNMqT/NztyPM15F9xytQhrBpelJWNV4uNvh7jO60SpBETgghhBBCDIGiKGSXNBIX7s5vb5yDlVbNX945xuHMqjPnNLV1k1FYz4LpXqgHMfohLMPL1RYVTInKld+vj+u/0MkPLZ3lS3ltW68fVPSmtaOHE0WNJER7DmrUb6QkkRNCCCGEEINWUddOa0cPkQHO+HvY8/DNCQR7O/Di5xl8ui8fk6Jw5EQVioJsAj7O6a00uDpaT4kRuYzCevw87HCy1w/6mrnRnuis1Ow7Xj6o85NzajApCnOixmbPREnkhBBCCCHEoOWUNAIQGegMgKOtjl9dM5vFM7z5fH8hL36WwYG0SgI97Ue9ap8YOW83WyomeSLX3WMkp6SJ6YOcVnmajV7L3GhPDp+oHlRBmMTsGtydrAnychhuqEMiiZwQQgghhBi0nJJGnOx1eDrbnHnNSqtm40UxXHleGIlZ1RRVtcho3ATh7XpqC4KhrAObaHLLmjAYTUOaVnna0pm+dHUbOZpV3e957Z09ZBbWkxA1NtMqQRI5IYQQQggxSKfXx0UFOJ/zZlWlUnHBgiDuvTyW6cEuLJohidxE4O1qS1e3kcbWoZXZn0gyC+vRqFVEBjgP+doIfye8XW37nV5pMil8sq8Ao2nsplUCaMfsTkIIIYQQYkKrbeqkoaWr3zfEsyM8mB0xdm9mxch4f7enWmVdGy4Og18/NpFkFjYQ5ueEtW7oqY9KpWLpTB8+3H2Siro2fNzOni7c0WXgpS8yScmrZUW8H6G+juYKe0AyIieEEEIIIQblzPq4YYxsiPHJZ5JvQdDa0UNxZcuwplWetmiGN2qVim+PV5z1em1jB39+O4nUk7VcvzqSG1ZHjtm0SpBETgghhBBCDNLp/eN8pYjJpOHsoEdnpZ60BU9OFDWgwJALnfyQk72eWeFu7E+vxGA0AZBb2sjjbyZS19zFA1fNYtUc/zFN4kCmVgohhBBCiEHKKWkkMsBZ9oabRNQqFd4utpN2RC6joB4bvZZgn5FVklw605fk3FrS8uto7zTwxldZuDpa8/MrZp4z3XKsSCInhBBCCCEG1NjaRVVDB+fN9rN0KMLMvN1sB73p9USTWVhPdKAzGvXIJiLGhrniZKfjza+yaWrrJjrQmbsvi8XexspMkQ6dTK0UQgghhBADkvVxk5e3qy11TZ30GAbeK20iqW7soLapk2kjmFZ5mkatZslMH5raujkvzpf/d3WcRZM4kBE5IYQQQggxCDkljeh1GgK97C0dijAzbzdbFKCqoQN/j8nz9c0sqAcYUaGTH7p4cQizwtwJ83Mc8/VwvZEROSGEEEIIMaCckkYi/JxGPEVNjD8+rqfWeFXWTa51cpmF9bg66vH+rjLnSFlp1YT7O42LJA4kkRNCCCGEEANo7eihtKZNplVOUl6uNgATrnJlbVMH+9MqKKtpxaQoZx0zmRROFDUwLch13CRe5iZTK4UQQgghRL9ySxsBWR83WVnrtLg46CfciNxrW7I4UdQAgI1eQ4iPI6G+ToT5OqJRq2jrNDAtxDzTKscjSeSEEEIIIUS/ckoa0WrUhIywhLsYv7xdJ9YWBCfLmjhR1MCFC4Lw+a7q5snyJrYcLDprdG5a0MgLnYxXksgJIYQQQoh+5ZQ0EurriJVWY+lQxCjxdrXlUGYViqJMiKmImw8WYWetZf2iIKx1WhbH+gDQ1W2ksLKZ/PJmdFYaHO10Fo509EgiJ4QQQggh+tTRZaCospULFwZZOhQxirzdbOnoMtDc3oPTOE9+SqtbScmr5dIlIVjrzk5n9DoNUYEuRAVO3imVp0mxEyGEEEII0aeT5U2YFIUoWR83qfl8V9mxsq7NwpEMbMuhIvQ6DSvn+Fs6FIuSRE4IIYQQQvQpp6QRtUpFmJ+jpUMRo+h0if7xvk6uuqGdwyeqWDHbz+IbcluaJHJCCCGEEKJPOcWNBHk7nDOFTUwurk7WWGnV4z6R23q4GI1azZq5AZYOxeIkkRNCCCGEEL3qMRjJr2iWaZVTgFqlwsvFhopxvAVBQ0sX+9MqWDrTB2d7vaXDsThJ5IQQQgghRK/yy5sxGBXZP26KGO9bEGw7UozJBOvmB1o6lHFBEjkhhBBCCNGrnJJGVEBEgJOlQxFjwNvNltrGTgxGk6VDOUdrRw97UsqZP80LD2cbS4czLkgiJ4QQQgghepVT0oifhz121lO7qMRU4e1qi0lRqG7osHQo5/gmsYSuHiMXLpDRuNMkkRNCCCGEEOcwGE3klcn6uKnE29UOGH+VKzu6DOxIKmV2hDt+HvaWDmfckEROCCGEEEKco7iqla4eI5GBzpYORYyR8boFwZ6Ucto6DVy0MNjSoYwrI07kWltbWb9+PaWlpecc++c//8mKFSu45JJLuOSSS/jvf/870tsJIYQQQogxkJJXg0qFFDqZQmyttTjZ6agcR5UrewxGth0pZlqwC6G+spfhD41oQ5DU1FQefvhhCgsLez2enp7O3/72N2bPnj2S2wghhBBCiDFkMinsT6skNtQNJzudpcMRY8jSlSsVRaGlvYfy2jYq6trILGqgqa2bOy6ebrGYxqsRJXIffPABf/jDH/j1r3/d6/H09HRefPFFysrKmDt3Lg899BB6vez5IIQQQggxnqUX1NHQ0sW1qyIsHYoYY95utiRmVY/Z/RpaukjPr+NkeTMVdW2U17bR1mk4c1yv07B0pg/RMsX3HCNK5J588sk+j7W1tRETE8ODDz5IUFAQv/nNb3j++ed54IEHRnJLIYQQQggxyvalVuBga0VchLulQxFjzNvVlrZOAy3t3TjYmn80tsdgIre0kfT8etIK6iiraQPAzlqLr7sdCdGe+LjZ4etui6+bHS4OelQqldnjmAxGlMj1x87OjpdeeunM3zdu3Mhvf/vbISVybm5SlcbcPDwcLB3CpCL9aX7Sp+Yl/Wle0p/mJ31qXuboz6bWLlLyavnJ0lB8vGX/uKn2PRoV4gbk0WmCUDM+e15JIy+8cojjebV0dRvRalRMC3Fjzfwg4qO9CPJ2kIRtiEYtkSsvL+fAgQNcccUVwKn5rlrt0G5XV9eKyaSMRnhTkoeHAzU1LZYOY9KQ/jQ/6VPzkv40L+lP85M+NS9z9ee2I8UYTQpzItyn/NdnKn6P2mhPJVNZJ2vxsDffiNy/NqVQVd/O4hnezAh1IzrQGWvd97lBbW2r2e41WajVqn4HtkZt+wFra2uefvppSkpKUBSF//73v6xevXq0bieEEEIIIUZIURT2Ha8gzNcRP3c7S4cjLMDdyRqNWmXWgifdPUYKyptZPS+IG9ZEERfuflYSJ4bH7Inc7bffTlpaGq6urvzxj3/kZz/7GevWrUNRFG655RZz304IIYQQQphJfnkz5bVtLJ3la+lQhIVo1Gq8XG3JKW3EaDKZpc2CimaMJoXpYW5maU+cYpZUeOfOnWf+/MN1cWvXrmXt2rXmuIUQQgghhBhl+46Xo7NSMzfa09KhCAtaNceft7Zl89a2bG5eFz3itWvZJY2ogGnBrnS0dZknSDF6a+SEEEIIIcTE0dlt4PCJauZGe2Kjl7eIU9mK2X40tHTy5YEiHGx1XL48bETt5ZQ04udhj72tThI5M5KfUiGEEEIIQWJWDV3dRpbOlGmVAi5bGkpLew+bD55K5tbMDRhWOwajiZNlzSyJ9TFzhEISOSGEEEIIwb7j5Xi72hLhL1sOCFCpVNy4JorW9h7e25GLg60VC6d7D7md4qpWunqMRATI95W5jVrVSiGEEEIIMTFU1LWRW9rE0pk+speXOEOtVnHHxdOIDnTm1c0nOH6ybsht5JQ0AhAZ4Gze4IQkckIIIYQQU923xytQq1QsmjH0ERcxuVlpNdx3+Uz8POx4/pM08sqahnR9TkkjXi42ONvrRynCqUsSOSGEEEKIKcxgNLE/vZKZYW44yZtt0QsbvZYHrorD2UHPsx+mUlYzuM27TYpCbmkjETIaNyokkRNCCCGEmMLSTtbR3NbN0llSjEL0zclOxy+vjkOrUfPy5hODuqa8to22TgNRksiNCknkhBBCCCGmsH3HK3Cy0zFTNmsWA/BwtuGihUEUVbZQWj3wqNzp9XEyIjc6JJETQgghhJiiGlu7OH6yjkWx3mjU8rZQDGz+NC80ahX70ysGPDenpBEXBz0eTtZjENnUIz+xQgghhBBTkElReG9HLiZFkb3jxKA52J4avT2YUYXRZOrzPEVRyClpJDLAWSqhjhJJ5IQQQgghpqBP9uZz5EQ1ly8PxdvV1tLhiAlkcawPzW3dpOfX93lOTWMHja3dRMq+hKNGEjkhhBBCiClmT0oZmw8WsTzOlwsXBFk6HDHBzAxzw97Giv3plX2eky37x406SeSEEEIIIaaQtPw63tqWw4xQV25YEynT3sSQaTVqFkzzIiW3hrbOnl7PyS1pws5ai4+73RhHN3VIIieEEEIIMUUUV7Xw/Kfp+HvY8bNLZkiBEzFsi2N9MBgVjmRW9Xr89Po4tXxQMGrkp1cIIYQQYgqob+7k/z5MxVav5edXzsJGr7V0SGICC/Syx8/DrtfplQ0tXVQ3dhDh7zz2gU0hksgJIYQQQkxyHV0G/u/DVLp6jDxw5SxcHPSWDklMcCqVisUzfMgvb6airu2sY7mljQBEBTqPfWBTiCRyQgghhBCTmMFo4vlP0qioa+fuy2Lx97S3dEhiklg43Qu1SsWBH43KZZc0orfSEOgl32ujSRI5IYQQQohJ7J1vcskobOCmdVFMD3a1dDhiEnGy1zMj1JUD6ZWYTMqZ13NLGgn3c5Q1mKNMelcIIYQQYpLaeayU3cllXLAgUDb9FqNi0QxvGlq6OFHUAEBrRw+lNW2y7cAYkEROCCGEEGISOlFYzzvbc5kV5sbly8IsHY6YpGZHuGOr17I/vQL4fn2cJHKjTxI5IYQQQohJprqxg+c/TcfL1YY7Lp6OWi0l4MXosNJqmDfNi2PZNXR0GcgtaUKrURHi42jp0CY9SeSEEEIIISaRji4D/9h0HID7r5gp2wyIUbd4hjfdBhOJWdVklzQS4uOIzkpj6bAmPUnkhBBCCCEmCZOi8NIXmVTUtfOzS2fg5WJr6ZDEFBDq64iXqy07k8soqmyRaZVjRBI5IYQQQohJ4tN9+aTk1XLt+RFMkwqVYoyoVCqWxHpTVNmCSVEkkRsjksgJIYQQQkwCe5NL+fJAEctm+bIy3s/S4YgpZuF0b1SASgXhfk6WDmdKkEnTQgghhBATXHFVC8++l0yEvxM3rIlEpZLiJmJsuTpaMzPMjfYug6zLHCPSy0IIIYQQE9z2xBKstGruuSwWrUYmXAnL+NmlM1CUgc8T5iGJnBBCCCHEBGY0mUjNq2PudG8c7XSWDkdMYVKpcmzJRzZCCCGEEBNYbkkTrR09LJjhY+lQhBBjSBI5IYQQQogJ7FhuDVqNmvgoT0uHIoQYQ5LICSGEEEJMUIqikJxTy4wQVykwIcQUI4mcEEIIIcQEVVLdSl1zJ7Mj3C0dihBijEkiJ4QQQggxQR3LqUGlglmSyAkx5UgiJ4QQQggxQR3LqSXCzwlHW6lWKcRUI4mcEEIIIcQEVN3YQWlNK7MjPSwdihDCAiSRE0IIIYSYgFJyagAkkRNiipJETgghhBBiAjqWU4O/hz2ezjaWDkUIYQGSyAkhhBBCTDDNbd3kljURHylFToSYqiSRE0IIIYSYYFLzalEUmB0h0yqFmKokkRNCCCGEmGCSc2txc7Qm0Mve0qEIISxEEjkhhBBCiAmks9tAekE9syPdUalUlg5HCGEhksgJIYQQQkwg6fn1GIwm4mVapRBTmiRyQgghhBATSHJuDXbWWiICnCwdihDCgiSRE0IIIYQws7KaVpKya8zersFoIjWvjrhwdzRqeRsnxFQm/wIIIYQQQphRj8HEcx8d51+fpHEwvdKsbeeUNNLeZZBNwIUQksgJIYQQQpjT9sQSaho78XGz5bWtJ8gpaTRb28k5tei0aqaHuJqtTSHExCSJnBBCCCGEmTS1dvHFgULiwt35nxvm4OZozT8/TqO6sWPEbSuKwrHcGqaHuKK30pghWiHERCaJnBBCCCGEmXy0Jx+DwcTVK8Oxt7HiF1fOQlEUnv0wlfbOnhG1XVjZQkNLF/EyrVIIgSRyQgghhBBmUVjZzP60ClYnBODlaguAl6st91wWS3VDBy98mo7BaBp2+4czq1CrVMwKdzdXyEKICUwSOSGEEEKIEVIUhXe+ycXB1or1i4LPOhYd5MJN66LIKGzgnW9yURRlyO23dfawJ7WceTGe2NtYmSlqIcREprV0AEIIIYQQE92RE9XklTbx0wuisbU+9+3V0pm+VNa3s/VQMd6utqyZGzCk9ncdK6Or28i6+YHmClkIMcFJIieEEEIIMQJdPUY+3J1HoJc9S2J9+jzv8uVhVNV38P6OXDxdbIgb5BTJ7h4j2xNLiA11I9DLwVxhCyEmOJlaKYQQQggxAl8dLqa+uYvrzo9ErVb1eZ5apeL29dMI9HLgpS8yaGjpGlT736ZV0NLew4ULZDROCPE9SeSEEEIIIYapvrmTrYeKmBvtSWSA84Dn63Ua7rp0Ogajwrvf5Ax4vtFk4qvDxYT5Og6qfSHE1CGJnBBCCCHEMH24+yQKcOWKsEFf4+Viy8WLg0nMriE5t6bfc49mVVPb1MmFC4JQqfoe7RNCTD2SyAkhhBBCDENuaSOHM6tYNy8QdyebIV27dl4g/h52vP11Dh1dhl7PURSFLQeL8XGzZVaEbDkghDibJHJCCCGEEEOUlF3Nsx8ex8VBz4ULgoZ8vVaj5uYLomls6eLjvfm9npOWX09pTSsXzA9CLaNxQogfkUROCCGEEGKQunuMvLktm399ko6Xqw0PXTcbvU4zrLbCfJ1YOcefnUmlnCxvOuf4lkNFuDjoWTDda6RhCyEmIUnkhBBCCCEGoaymlcffTGR3chnr5gfyPzfMwdPFdkRtblgWirODnje2ZmEwms68nlfWRE5JI2vnBaLVyNs1IcS55F8GIYQQQoh+KIrCnpQyHn8jkea2bh64ahZXrQg3S4Jlo9dyw5pISmva2Hak+MzrWw8VYWetZdmsvvelE0JMbbIhuBBCCCFEH9o7DbzxVRZHs6qJCXLh9p9Mw9leb9Z7zI7wYE6UB5/vLyQh2hODUSE5t5aLFwdjrZO3akKI3sm/DkIIIYQQffjPFxmk59dz+fJQLlgwekVHrjs/kszCQ7z5VTYuDnp0WjWr5viPyr2EEJODTK0UQgghhOhFYWUzx0/WcdmyEC5aGDyqlSNdHPRccV44J4oaOJBeydJZvjjY6kbtfkKIiU8SOSGEEEKYncFo4rNvC8gpabR0KMO25WARNnoNK2aPzcjY8jhfwv2dUKtUrJ0XMCb3FEJMXDK1UgghhBBmZTCaePGzDJJyaiiqbCEywHlM7ltc1UJVQweBnvZ4uNiMaAStoq6NpOwaLlwYhK312LxdUqtU/PyKmdQ2dg55g3EhxNQjiZwQQgghzKbHYOKFT9NJyavF1VFPYWXzmNw3vaCO5zYdx2BUALDWaQj0tCfQ24Egr1P/+brboVYPLrnbeqgYrVbN6oSxHRmzs7bCzttqTO8phJiYJJETQgghJhGToozqWq7+9BiM/PPjdNLy67hhTSRGk8K73+TS0NKFi4N5Kz3+UHZxA//8KA0fNztuXBNFeV0bRVUtFFe1sDelnG7Dqf3ZIgOc+dU1cQNuG1DX1MnBjErOm+2Ho52sUxNCjE+SyAkhhBCTxMnyJp7bdJwrzgtj6UzfMb13V4+Rf350nMzCBm5eF8XyOD/ySpuAU0VDXBw8RuW+J8ua+L9Nx3FzsuaX18ThaKsj3N/pzHGTSaGivp2U3Bo+2pPP5/sL2LAsrN82T+/ntm5e4KjELIQQ5iDFToQQQohxqLWjB0VRBn1+ZX07z354nJb2Hj7YmUdrR4/ZYjmUUcnB9Erqmjp7Pd7VbeTZD1PJLGzglgtjWB7nB0CAlz1qlYrCihazxfJDRZUt/O2DVJzsdDx47Wwce6nyqFar8HO346KFwSyJ9WHzgSKyihr6bLO5vZu9qeUsmO6Fm5P1qMQthq7T0MnWgm+o6+j7ayfEVCMjckIIIUaFoijUNHaQU9KEVqNiwXRvS4c0YWQW1vPX91OYF+PFxgujsdJq+j2/sbWLv72fgkoF926I5V+fpPHpvnxuWBM14lhqGjv4zxeZZ/7u5mhNZIAzUYHORAY442Sn49kPU8kta+K29dNYOOP7r7PeSoOvux2FleZP5EprWvnr+ynY6rU8eM3sQW3Sfd3qCHLLmnjpy0we2zgPe5tz16J9k1hCj8HEhQuCzB6zGB5FUXgv+xOOViWzu3Q/t824gQiX/kdVhZgKJJETQghhFiaTQmlNKzkljeSUNpFb2khTa/eZ4zFBLjgN4s32VNfRZeC1LSewt7HicGYVtU0d3LthJk59rNVq7zTw9w9SaWnv4dfXzSbEx5EVs/3YlVzGeXF++HvajyieQ5lVADxw1Syq6tvJKWkko6COgxmVAGg1akwmhTsvns68GK9zrg/2cSA1rxZFUVCZae1eZX07z7yXglaj4sFr4wY9cmat03LXxdN54s1EXttygns3xJ4VU0eXgR1JZcRHeeDjZmeWWMXIHa5M4mhVMkv8FpDbcJLnUl7iyoiLWeq30GzfU0JMRJLICSGEGLE9KWV8sOskHV0G4NTmxtGBLkT6O2Fvq+OFT9NJL6hncayPhSMd/97fmUt9Sxe/vXEODc1dvPxlJk+8kcjPr5yJv8fZSVmPwcQ/Pz5OeW0bP79iJiE+jgBcujSUw5lVvPNNDg9eO3vYb3YVReFQRiWR/k7EhroRG+rG+QkBKIpC5XdJXUFFC/GR7swMc++1jRBvB749XkFds3lK6tc0dvD0u8mgKDx4XTyeLrZDuj7I24Erzgvj/Z157E4uY0X893vE7Uouo6PLwEULZTRuvKhqq+b97E+IcA7l6shL6TJ28XrGu7yf8yklLeVcFXUpVurhvZ0taSnnw5zPuCz8QkKc5GsuJh5J5IQQQozIybIm3v46hzA/J5bH+RLh73TWG3ZFUXCy05GWXyeJ3ADS8uvYm1rBhQuCCPN1Al9wc7LmuY+O86e3krjrkhnMDHMDTlWnfPnLTLKKG7ltfQwzQt3OtGNvY8Vly0J5++sckrJrSIj2HFY8xVWtVNS1s3rt2VM0VSoVPm52+LjZsTyu/zaCv0suCytazJLI/eeLDLp7jDx0XfywR81Wzw0go6Ce93bmERHgjL+HPd09Rr4+WsL0YBeCvR1HHKcYuR5jD69mvIOVxoqfTr8WtUqNjdaGO2f+lC/zv2Zb0U4q26u4bcZNOOkdhtR2Rl0Wr6S/TZexm10l30oiJyakERc7aW1tZf369ZSWlp5z7MSJE2zYsIG1a9fyu9/9DoPBMNLbCSGEGEdaO3r492fpuDjouf/yWBZO9z7nzbpKpWJGqCsZBfUYTSYLRTr+tXf28PrWLPzc7bhkSciZ10N8HHnkpgQ8nW14dlMq3ySWnFoz9E0uR7OquXJFGItmnJsgL4/zxd/Djg925dHdYxxWTAczKtGoVcNOBAH8PezRqFVmWSfX3NbNybJm1s4LHNGUUbVKxa3rp2Gj0/Di56cSw2/TKmhu6+aihcEjjlOYxycnt1DaWs6NMVfhrP++EqlapebisHXcOuMGSlvKeSrxOYqaSwbd7r6yg/z7+Ot42rgz23Mmx2sz6TR0jcYjCDGqRpTIpaamcu2111JYWNjr8QcffJDf//73bNu2DUVR+OCDD0ZyOyGEEOOIoii8uvkEja3d/OzSGdha972JcWyoG22dBgrKR6d64WTwzje5NLV2c+v6GKy0Z/96dnW05jc3xDMrzJ13vsnlz28f45ukUtbMDeizRL5Greba8yOpbeo8U05/KEwmhcMnqpgZ5tZrUZDBstKq8fe0N8vG4JmF9QBMD3EdcVtOdjpuXT+Nspo23tuRy1eHiwnzdSQq0HnEbYuRO16TwZ7S/azwX0Ks+7Rez4n3nMkv59yDWqXmb8deYFPO59R21PXZpkkx8UneZt7L/oRprpH8Iv5nnOe/mB5TD8drM0brUYQYNSNK5D744AP+8Ic/4Ol57id1ZWVldHZ2EhcXB8CGDRv46quvRnI7IYQQ48jXR0tIyavlqpXhZ9Zm9WV6iCsqFRzP7/tN1lSWnFvDgfRKLloY1Oe0Pmudlns3xLJufiB5ZU3Mi/HkqpXh/a5/iwlyYU6UB5sPFVHf3PvWAX05UdxAU2u3WaqNhng7UFjRMqTtFHqTUViPnbWWIK+hTaPrS2yoG2vmBrA7pZzapk4uWhgsxTPGgYbORt4+8SEB9r5cEn5hv+f6O/jyUML9xHvOZE/ZAR49+BT/SXuTvMaCs77fuo09vJr+X74p3sMyv4XcEXsz1lo9oU5BuOidSapKGeWnEsL8RrRG7sknn+zzWHV1NR4e32/+6eHhQVVV1UhuJ4QQYpw4WdbEpt0niY/04Pw5/gOeb2dtRZivE+n5dWxYFjoGEU4crR09vPFVNgGe9vxkcXC/56rVKq5aEc6SWB+8XG1QDyLpuGpFOKl5dWzafZI7Lp4+6LgOZVRirdMwK8xt4JMHEOzjyO6UcqobO/AaYnGS0xRFIaOgnmnBrqjV5ku2Ll8eRm5pI4oCM8NH/qxiZEyKidcz36VHMXDLjOsHVcjEXmfHzdOu4ZKwC9hbepBvyw6RWpNOoIMfKwKWEukSxstpb1HYXMKG8PWsDFh6JmFXq9QkeMWxo2QvrT1t2FtJtVIxcYxasROTyXTWp1rDKTvs5jayksniXB4e5vkUU5wi/Wl+0qfmNRr92dLezX++zMTN2YYHb5o76Gl3C2J9ePurLKysdTg7TMxtCEajP19/O5G2jh6euGsRPt5OA18wxDg8PBy4fEU473+Tw2UrI5gWMnCy0tVj5FhOLUtm+eHn6zzoe/Ulfpo3r2/Nor6thxmRZ8c+2GcpqmymsbWbBTN9zf51+PsD52EwKeit+t+vbyKY6P+Gfpj+JXmNBdwz72ZmBA3tQx8PHIjwv5IbDJewt/AwW3J28kbme6hQYaXR8svFdzDPP+6c61ZrF7G9eDd57bmsDl96brsTvE/HG+lP8xm1RM7b25uampozf6+tre11CmZ/6upaMZlGNg1DfM/Dw4GaGlmfYi7Sn+YnfWpeo9GfiqLwj4/SqG/q5Lc3zqGjtZOO1sFN2Qv1PvXh3J7Eol6Lc4x3o9GfiVnV7E0u49KlIdhbqUft+/+8mT58fbiI5z9M5ZGbEwYc0TqaVU1Hl4G4MFezxGStPrVW7nhONTH+3yerQ+nTb4+dKqoW6GYr/070YaL/G1rWWsGmjC3M9Ypnmv30ET1LnFMcMxNmcqI+h6SqVJb5LyRYH9hrm7aKE162nuw+eYg4p7izjk30Ph1vpD+HRq1W9TuwNeKqlX3x8/NDr9eTlJQEwGeffcayZctG63ZCCCHGwFDWxf1YoJcDjrZWpOXXj1J0E0trRw9vfZ1NkLcDFy4Y3dLnep2GK88Lo6iqhZ3Hzq0y/WMH0ytxstcRHehilvtrNWoCPe0pqBj+G7iMgnq8XW0Hvfm3mHj2lx9Bo1JzZeTFZmlPrVIz3S2am6ZdTbBj70WB4FRl3QSvWeQ1FtDQ2WiWewsxFsyeyN1+++2kpaUB8Mwzz/DnP/+ZdevW0d7ezk033WTu2wkhhBgDiqJw/GTtkNbF/ZhapWJ6iBsZBfUy2wLYkVRKS3sPt1wQjVYzap+rnjF/mhexoW58sCuPgoq+K0i2dvSQll/HgmleZl2LFuztSFFVy7C+9j0GE9klDUwPHnm1SjE+GUwGEquSifWYjp3V8NZRjkSCVxwKCknVqWN+byGGyyy/OXbu3Im//6lf6i+99BKxsbEAREdHs2nTJr766iv++te/otPpzHE7IYQQY0RRFFLyavnz28f4vw+P4+5kzcYLo4dd2S82zJXWjh4KzFCKfiLr6jbyTWIJceHuBJqpAuNAVCoVt/9kGk52Op7/JI3Wjp5ez0vMqsZoUlgwbeTVKn8o2MeBrm4jlfXtQ742r6yJ7h6TWbYdEIPTY+xhd+l+Xkp7i+bu0Z8Kl16XRVtPOwu854z6vXrjaetBoIO/VK8UE8qorZETQggxcRlNJo5mVbPlYDGlNa24OVpzw5pIlsT6oBtBQYgZIW6oVJB2so4w38EV9piM9qaW09ZpGPUplT9mb2PF3ZfF8ue3k3jpi0x+fuXMcypfHsyoxMfNlkAv8xYcC/5uKm5hZTO+7kOrDJhRUI9GrZI93sZAj8nAwfIjbCvaRWNXEypU1HfW8/PZd2GtHb0iRYcrknDUORDjGjlq9xhIglccH+d9SVV7DV62HgNfIISFjf5cDiGEEBNCj8FIQ0sXu1PK+O1/DvGfzzMxmkzctj6GP9+5gJXx/iNK4uBUIhHq4zil18kZjCa2HS0m0t+JcP+xT2ZDfBy5dlUEafl1bD5QeNax2qYOckubWDjd2+z7qfm42qK30gxrnVxGYT1hvo7Y6OXz59FiMBnYV3aQxw4+xfs5n+Jq7cJ9cbdz18yfUtpawcvpb2E0GUfl3i3draTXnWCu12w0astVDp3jNQsVKhJlVE5MEPIvohBCTDHHT9ayJ+XUiFB7Zw9tnQbaOnroNpjOnBPi48jVKyOIi3Af1F5lQzEj1I3Pvy2gub0bR9uJP+W+x9jD4cok5nvPwUoz8FYMhzOrqG/u4qa1UWMQXe/Om+1HbmkTn+4rINTP6czas8OZp/Z7XTDNy+z3VKtVBHnZUzjEabXN7d0UV7Zw6dIQs8ckTo3AHalIYmvhDhq6GglxDOL6mCuIdok4k8xfG7WB/2Zt4r9Zm7gx5iqzJ/mJVSmYFBPzfSwzrfI0Z70T4c4hJFWlcGHw+eNqc/jm7hZstTZoB7Gvnpg65LtBCCGmEEVReH9nHs1t3fh52OPhbEOwjRX21lbY2Wixs7bC192OCH+nUXsTExvqxmffFpBZUM+C6eZdh2UJe8oO8EneZjoMnawOOq/fc02KwpZDRfh72BEbarnNp1UqFTevi6a4upUXP8vg0Vvm4uKg52BGFeH+Trg724zKfYN9HNmVXIbRZEKjHtykoBOFDSjANFkfZ1bV7TV8W36YwxVJtPa0EewYyHXRlxPjGnnOz/4i33k0djWxuWA7znonLg5bZ9ZYDlckEuDgh5+95bclSfCK493sjylpLSPQYehFnUZDc3cLfzz0DCFOgdw9c+O4SjCFZUkiJ4QQU0hJdSsVde3cuDaKFbP9LBJDsI8D9jZWpyojTvBErtvYwzfFewD4pngPS/0W9ruOKDW3loq6du74yTSLvxnT6zTcc9kM/vhGIi98ls5150dSXtvGjaM4Uhjs40DPURPlte0EeA5uDV5GYT22ei0h3kPb7kKcy2AycLw2k31lh8hpyEOtUjPTfTpL/RYQ5RLe7/fkBcHn09jVzLainTjrHVnmv8gsMZW1VlDSWs6VEZeYpb2RivOM5YOcz0isShk3idxnJ7fSYeggsy6b47UZzPKYYemQxDghiZwQQkwhh09UoVapmBNluYX8apWKGaGupBfUY1IUs0/dHEsHKo7Q0t3KJWEX8NnJrewrO9jnqJyiKGw+VIS7kzVzYzzHNtA++LjZccsF0fz7swye3XQcjVrF3OjRi+10MlZY0TyoRE5RFDIK6pkW7GLWrRCmmg5DB18X7eZg+VFaelpxtXbhJ6HrWOiTgJN+cAmySqXi6shLae5u4YOcz3DUORDnGTvi2A5VJKJRaUjwihtxW+Zgb2VHjGskSVWpXBp24bDaUJRTW2yY48OagqZiDlUksjJgKVn1uWzK/YIY1yh0g5jGLSY/SeSEEGKKUBSFI5nVTAtxsfjatNhQNw5lVFFU2TLkjcXHC4PJwPai3YQ5BbM68DxyG/L7HZXLKWkkv7yZG9ZEDnpa4ViYF+NFXmkT3ySVEhfujr3N6L1B9HCxwUavpaCyhaWzBj6/oq6dhpYumVY5Qu9lf0JSVSqx7tNY4jefGNdI1Kqhfw9q1Bo2Tr+O55Jf4rXMd7lPZ0+48/DXLhpNRo5WJTPDPQZ73dAqmY6mBK840utOcLKxEC/PQXyj/kBbTzt/OvJ32nvacdQ54Kh3xFHngJPe4dT/dY7MHOReeSbFxIc5n+Gkc+CikNXMdJ/G/yW/yNdFu1gfuma4jycsIK02k4/zvgRAr9ah05z6T6/RodfocdI7siZoBTZa6yG1O35+kwghhBhVJ8ubqWvuZH6M+QtZDNX0EFdUQFp+naVDGbYjlcdo7GpiXfAqmtu6uSB4Fa09bewrO9jr+ZsPFeFga8WSWMuvA/qxq1aGs25+4KgXFFGrVAR7O1DYz4bkP5RReKq6qWwEPny5DfkkVqWwLngld868melu0cNK4k7TaXTcNeunuFm78O/jr5PXWDDstjLrs2npbmW+hfaO60us+zR0aisSq1OGfO2Wgu00dTWz0HcuwU6BaFUaKtqqOFKZzBf523g760OeTX6RDkPngG0dqkikqKWES8MvwlprTYRLGAlecWwv3k1tx8T9t3OqKWwu5pX0/6JWafC398VJ74hapaa9p4Oq9hryGvPZXrSbf6W8TIehY0hty4icEEJMEYczq9Bq1MRHWn5/JEdbHcE+DqTl13Hx4olXjdBoMrKtaBeBDn6Ymt154P39zApzIyIkvNdRueKqFtLz69mwLHTEWziMBq1GzVUrwsfkXsE+Dnx9pISeH1RJ7UtGQT2eLjZ4jFLxlcnOaDLyYe5nuOidWRO0wmzt2lvZcc+sW/ln6sv8I/k/XBt9OQt8EobczuGKJOyt7JjuZrkKrr2x1uqJdZ9GcvVxDEPYcqGirYq9ZQdZ7DefqyIvPed4t7GHE/U5vJz+Fq+kv83PZt7S53YL7T0dfHZyK6FOwcz1mn3m9cvCLyKtNpNNuZ9z18xbhvxsYmzVdtTz79TXcdQ58IvZd+Kg631KeUpNOq+m/5fnkl/ivrjbsB3EiC3IiJwQQkwJJpPC0axqZoW5jZu9uGJD3cgvb6a1o8fSoQxZUnUqtR11rAtexa5jZdjoNZwobiA70ZPWnjb2lB446/wth4qw1mlYGW+ZAjPjSYi3I0aTQllta7/nGYwmsosbmS7TKodtf/lhylor2BCxHp3GvNOp3WxceXDOvYQ5h/DWiQ/4NG8LJmXg5Py0tp520mozmes1e1yW1E/wiqOtp51j5WmDOl9RFD7K/QK9Rsf6kN6nPeo0VszymM61URs4UZ/Du9kfn1lP92ObC76mraedqyIvOWutnbPeiQuCzyet9gTptSeG/mBizLT3tPNC6qsYFCN3z9rYZxIHEOcxg9tjb6S8tYJnk/9Da3fboO4hiZwQQkwBWcUNNLd1M28U9gcbrthQNxTl1KjLRGJSTGwr3ImvnTf++jBS8+o4L86PJ26dT6RrCMZGd77I3UFe+ampT9UN7RzNqua82X7YWkuBgmBvB4ABNwY/WdZEV49RplUOU2tPG1/kbyPSOYzZHiMvStIbWytb7pl1K0v8FrC9eDcvp71Fp6FrUNcmVaVgUIwW3zuuL9PcovC0deelpHdp7Goa8PyMuixO1OdwYcjqft+ww6ntHC4IXsXBiqN8VbjjnOPlrZVnRvYCHM798GdFwBK8bD35MPdzeowT74OwqcBgMvCftDep6ajjztib8LYbuIhUrPs07pj5Uyrbq3k2+UVauvv/sAskkRNCiCnhyIkq9DoNM8Mst3fZj4X4OGJnrZ1w6+RSazKobK9mbfBKDqRXYVIUls3yxd3Zhl9cOZP1oatRNN08vf0TPtyVxxcHCtGoVayZG2Dp0McFNydr7G2sBlwnl15Qj1qlIjrQZYwim1y+yN9Gp7GLK380omNuGrWGayIv48qISzhem8nfj71AQ2fjgNcdqkzCz94Hf3vfUYttJLRqLbfPuIlOQxcvp72NwWTo81yDycBHeV/gZevBcr/BbctwUcga5nvP4cuCrzlUkXjmdUVR+CDnU2w01vwkdG2fsV0VeQm1HXV8U7x3aA8mRp2iKPw3axO5jfncEHMlES5hg752ulsUP5t5CzUddfxf8ou0DDAyJ4mcEEJMcgajiaTsGmZHuKMfR+uz1GoV00NcSc6t5csDhRRUNGMy9T7NaLxQFIVthTvwtHEnziOWvSnlxAS54OV6aj2DSqVifdxsIp0jsPYvYuvRfPanVbI41gdn+773l5tKVCoVwT4OFFb2PyKXUVBPqJ8jttbjb9rdeFfSUsb+ssMs81uIr/3o79WoUqk4L2AxP5u1kdqOep5K/AeFzcV9nl/ZVkVRcwnzvedYfD/F/vjae3P3vBspaC7io9wv+zxvb+kBqttr2RC+vs81bz+mUqm4LvpyolzC+W/WJrLqcwE4Vn2c3MZ81oeuxd6q70qe0a4RzPaIZVvRTuo6Gob2YGJUbSnYzpHKY6wPWcM87/ghXx/tGsE9szZS39nAqxlv93uuJHJCCDHJpRfU09ZpGBfVKn/swgVBeDhZ8/HefB5/I5GfP7eP5z9NZ09KGbWNQ6veNRYy6rIoaS1nTdAKsgobqWvuZHncuSMKF4etxaDqYtVaA/NiPFm/MHjsgx3Hgr0dKatpo6un90ISrR09FFW2yLTKYVAUhQ9zPsPOypaLQlaP6b2nu0Xxq4R7sFJb8bekF3j22ItsK9xJUXPJWevnDlUkoVapmes9u5/WxoeFAXNYFbCMvWUHOFyRdM7xlu5WthR+wzTXKGa4xwypba1ay+2xN+Jt68lLaW9R0FTEx3lf4m/vyxK/+QNevyFiPSrg47wvhnRfMXoOVSSypfAbFngnsC541bDbiXAJ455Ztw44vVI+5hJCiEnuSGYVdtbacVk0ItDLgUc3zqO5rZvMonoyCurJLGwgMasagDBfRx68dva4qPSoKApfFe7ARe/MPO94/v1pJvY2VsyOOLcKaIhTINPcokhtPsIfL1rV675yU1mItwMmRaGgrAk3u3PXDWYW1qPAuPyeHe8Sq1I42VTIddGXD7rynTn52Hnx64T7+Lp4F1n1uXye/xWf53+FrdaGSJdwol3DOVJ5jGmuUTjqHMY8vuG4JOwCiltKeTf7I3ztvc9at/Zlwdd0Gbu5PGL9sNq20dpw96yNPJ34T/527AVMiolbpl83qC0iXK1dWBu8ii/yvyKpKoV4z1njeoRzsstrLOCdrI+IdAnn2ugNI/5ahDuHcPO0a/o9RxI5IYSYxLp6jCTn1jJ/mhdazfidhOFop2PBNG8WTPNGURTK69pJyqrm028LOJBeyXmzLV/tMaM6m4LmYq6OvIzWdgMpebWsTgjAStt7v14YvJpnkv7J3rIDZi39PhkEf7cJ/P7j5UT5nfqzosDpibWHM6uw0WsJ8ZkYb/THi86eTj7J20yggx8LfeZaLA57nR0bwk8lNs3dLeTU53GiIZes+lxSak5VgbzS5xKLxTdUGrWGW2fcwP8efZaX0t7kobk/x87KlrLWCvaXHWa5/yK87YY/48HF2pm7Z23k78f+zSyP6UPaZH1V4DISq5J5NeMddpTs4/zA5cxynz7oKZ7CPNp7Ong9411crJ25fcaNZqvE2luxmx+SRE4IISax1LxaunqMzI8ZuGLWeKFSqfBzt8N3cTDJebVsO1rCsjhf1Bb+pPmjzK046RxY6JPA10fKMJoUlvUyrfK006Ny3xTvwdvWEy87T9ytXeUNFuDioMfNUc+ne072eU5CtCca9fj98GE8+vjEVzR1N3Nb7I0j2vTbnBx1DiR4zybBezaKolDdUUtVW/WQpyFamoPOnttm3Mj/HXuB1zLe4e5ZG9mU8zm2WhsuNMMUVn8HX55Y/D/oNUMbvbdSa/l1wn0cqkhiZ8leXkl/GzdrV1YGLGWh71z0Zt52QpxLURTey/6Ypu5mfjnnbmytxm7fS0nkhBBiEjucWYWTvY6oCVj5T6VSsXZeAP/5PJPjeXXERbhbLJaCpmIyqnO+K2agZW9qOdGBzni79j91bX3IGv7v2L95Me0NANQqNR427njbeuBp64G/vQ/xXrPGzZvusfTQdfF0Kyoam9pRAahUqE79D4AATxmNG4qq9hq+zN7BfO85hDoFWTqcXqlUKrxsPfCyPXc68kQQ4hTIlZGX8G72x/wz5WVyGk9yVeSl2JlpCquNdngJgE6jY5n/Qpb4zed4bSY7ivfwYe5nbC74mmV+C1kZuMxsMYpzHak8RlJ1Kj8JXUuwY+CY3lsSOSGEmKBS8mppbutmSawPavW5o1VtHT2k5ddx3my/Xo9PBAlRnmxyPMm2I8UWTeT2lx9Gr9Wz2HceJ4oaqGns5LJloQNeF+QYwJ+WPExlWzVV7TXf/9dWTXpdFkbFSG1n/YgWxU9U7s42eHg4UFPTf/VKMbCW7lZePP46Oo0Vl4RdYOlwJrXFvvMpbC7hYMVRfOy8WOI7cFGSsaJWqYnzmEGcxwzymwrZUbyXbUW7KG+r4s6ZN1s6vEmptqOeD3I+Jcwp2CJT6CWRE0KICehYTg3/+iQNRYGdx0q5cU0UYX5OZ51zMK0Cg1EZl9UqB0urUXP+nAA+2JVHYWUzwd6OYx5Dp6GTpOpUFgcmYK21Zk9KHvY2VsyJHNyogo3WhhCnIEJ+NEpiNBl5Kf0tthftZrHv/AE3ERaiNx2GDv6V8jL1nQ08fN79ODH2PyNTiUql4urIS7HW6JnnHT9up0qHOgUTGhvMx7lfsrt0P+097RYpfjOZGU1G3sh8F1Bx87RrLTKzYurN5RBCiAkus7Cef3+WTqiPI7evn0ZzWzdPvpXE61uzaO3oOXPevpQy3J2sCfWd2G/sls3yxVqn4esjJRa5/7Hq43Qbu1kZuoimtm6Sc2pYNMMbK+3I3sBp1BouDbuALmM3XxXuGNK1PyzlLqaubmM3L6S+TllbJbfH3kSMR4SlQ5oSrDRWXBF5MYGO/pYOZUBzvGZhVIyk1mZaOpRJZ1vRTvKbirg26jLcbCyzfEESOSGEmEAKKpr5x8dpeLna8vMrZ7FwhjdP3r6ANXMD+PZ4Bb/9zyH2ppbT1NZNSm4N82K8Jnw5altrLctm+XLkRDX1zZ1jfv8D5UfxsvUk0i2UA2kVGE1Kr3vHDYe3nReLfOexr+wQNe11g7qmrLWCh/Y9RnrtCbPEICYmg8nAS+lvkd9UyE+nXcN0t2hLhyTGoUAHf9ysXTlWlWrpUCaV/KYithbuYK5XPAkW3A9REjkhhJggymrb+PsHqTjYWPH/rorD3ubU/ls2ei3XrIrg0Vvm4utmy+tbs/jDq0cwmRTmT5u40yp/6PyEU598f5NYOqJ2Wtq7+epw8Vkjl/2pbKuioLmIRb5zURTYk1pOZIAzPm52I4rjhy4KWY1GpeaL/K8GPLfL2M0r6f+l3dBB2hh+wt5j7JFRwHHEpJh4M/N9MuuyuTZqA3O84iwdkhinVCoV8Z4zyWrIpbWnzdLhDMuhikT+e+JDFEUZ+OQf6DJ2U95aafZ4OgydvJ7xLs56J66Osuw2GpLICSHEBFDb2MHf3k9Bo1bxq2vicHE4t0S1v6c9D10fz60XxYCiEOrnhL+H+RIOS3J3siEh2oM9qWV0dBmG3c6Hu0/ywa48Hn7pEEezqgd8Y3Cg/ChqlZr53nNIO1lLdUOH2UbjTnPSO7IqcBlJ1akUNhf3e+6HOZ9R3V6Dm7UreY0FZo2jL+097Txy4M9sK9w5JvcT/Ttd6jypOpVLwy5ksd/4KbYhxqd4r5mYFBPHazIsHcqQmRQTmwu2c6DiKAXNRUO69t2sj3nyyN/YWvCNWT+I+jDnM+o7G7h52jXDrjRqLpLICSHEONfU1s0z76fQ1W3kl1fH4enS94J1lUrF4lgf/vKzRfz57sUTflrlD62dF0hHl5F9qeXDur6iro39aRXMjfbExdGaFz5N558fp9HQ0tXr+QaTgcOVScS6T8NBZ8+2Q0XYWWtJiDJ/6fTzA5djb2XHp3lb+kwuEyuTOVhxlLXBK1nsO4/K9mpau0f/E/YdxXtp6WllT9kBDKbhJ9Fi5BRF4dOTW9hffoQ1QStYHXSepUMSE0CAvR/uNm4cqz5u6VCGLK8xn/rOBgB2FO8b9HV1HQ0kVafgrHfiy4KveSX9bToNI5+af6z6OIcrk1gXvHJIG7ePFknkhBBiHGvv7OHv76fQ2NrFL66chb/n4Cob6q002FpbjXJ0YyvEx5FIfye2J5ZiNA3909VP9uajs9Jw/ZpIHr5pDletCCe9oJ6HXz7M3tTycxKo9NoTtPa0McNxFoczqziYVs6iGT4jLnLSG2utNReGrCa3MZ+Muqxzjte01/Fu9seEOgVzYfD5hDuf2vrgZNPojsq1drexq/Rb3G3caOlu5bgUTBgzHYZOippLOFJ5jC/zt/FK+tv8+ej/8U3xHpb6LeTi0HWWDlFMEKenV2Y35NHS3WrpcIbkYEUiNlprVvgvIbUmndqO+kFdt6vkVNL3qzn3sCF8Pak1GTyT9K9Br0XuTWtPGx9kf0qggx8XBJ8/7HbMSRI5IYQYp0yKwgufplNW28a9l8US7u808EWT3Np5gdQ1d5KUXTOk6worm0nMrmFNQgCOtjo0ajXr5gfyx1vnEehpz+tbs3jmvRTKalrJLm5g66Ei3kneCT16Xn63hhc/z8BGb8XKeL9RejJY4jsfTxt3Pj255axpQAaTgVcz/otKpeaW6deiUWsIdPRHq9aO+vTK7cW76Tb2cGfszbhau/Bt2aFRvZ+ApKoU/ufbx/nV3t/zVOI/eCPzPb4q3ElxSxlOekcuDl3HVZGXTKrRdjH65njOwqSYSK1Jt3Qog9Zh6CS5Oo05nrM4P2g5KpWK3aXfDnhdW087+yuOkOAVh4u1M6sCl3Fv3G00dTXzVOJznKjPGVY8m3K+oM3QzvXRV46bbSdkHzkhhBindh0rI6OwgZvWRjEj1M3S4YwLs8Ld8XSxYduRYuZGew76zexHe/Kxs9aydl7gWa97udjy4HWz2ZtSzge78njklSOnDlh1YhNXhmdPLEtWRxPm58jsaT401I/eVEaNWsNPwtbxSvrbHKpIYpHvXAA+P/kVxS2l3B57E67Wp0pcW6m1BDsGjGoi19TVwp7SAyR4zcbX3pvFvvP5Iv8rqtpr8LI1//RScco3xXuxUp/a2NvL1gMvWw/cbNywUstbNjF8fvY+eNq6c6z6OEv8Flg6nEFJrj5Oj6mHBT4JOOudmOM5iwPlR7goZHW/a9P2lR2k29jN+YHLz7wW7RrBQ3Pv58Xjb/CvlFe4NPxCVgUsG/TvkPTaExytOsYFwavwdzDvOumRkBE5IYQYh6oa2vlwdx4zQl3NXlxjIlOrVayZG0BBRQu5pU2DuiarqIGMgnouWhiMrfW5b4bVKhXnzfbjidvmc/XKcH5+xUwuuVgLKvjZsgtYNcefYG9HtJrR/5U52yOWYMdAvszfRrexm4y6LHaU7GWZ30LiPGacdW64UwilreVmWffRm+1FuzAqRi4MWQXAQp+5qFVq9pcdHpX7CajtqKO4pZSlfgtYE7SCWR4z8LbzkiROjNip6ZWzyGk4SXN3i6XDGZRDFYl42XoS7HjqA7iVAUvpMnZzoPxon9f0GHvYXbKfaa5R+Nn7nHXM3caNX865h1keM/gkbzNvZL4/qHW/HYZO3s3+GG87L9YGrxrZQ5mZJHJCCDHOmEwKr2w+gVat5pYLYmQK1Y8sjvXBzlrLtiP9V3iEU8UhPtp7EhcH/YDTIl0drVk7L5DYMFeO1RwjwjkUT1t3c4U9KCqVisvCL6Kpu5lPT27lzcz38bP3YUP4+nPODXcOxaSYKGgauB+GqqGzkX3lh5jnHY/nd6NvTnoHZnnM4FBFIj3GwW3fIIbmWNWpYhTxnrMsHImYjOI9Z6KgkFI9/qdXVrfXcLKpkAU+c878Dgx09CfCOZRdJd9iNBl7ve5wZRItPa2sDlre63FrrZ7bZtzA+pC1HK06xr+Pv06XsbvfWD47uZWmrmZuiL5i3H2oIomcEEKMM18fLSGvtInrVkf0us3AVKe30rAi3p/k3Fq+PlLc7xYCqXl1nCxr5ieLg9FZDW5NQ15jPrWd9Sz0mWuukIck3DmEWPdp7CndT7exm43Tr8NKc27hmhCnQFSoyBuFgifbinZhUkznLOhf4jufNkM7yTVpZr+ngGPVqYQ4BuJm42LpUMQk5GvnjbetJ8eqx//m4IcrklChYp53/FmvrwhYSkNXI6m1526lYFJM7CjZS6CDHxHOYX22rVKpuCBkFddHX0FWfS7/THmZ9p6OXs/NbTjJvrKDrAhYQohT0MgeahRIIieEEONIWW0bH+/NZ3aEOwune1s6nHFr/cIg5kR68N7OPN76OgeD8dwqliZF4eO9J/F0sWFJrE8vrfTuQPlRrDXWzPaMNWfIQ3Jp2AU46Ry4JmoD3na9b+purbUmwMGXk2ZeJ1fX0cCB8iMs8pmLu43rWcciXcLwtHGXoiejoLq9hpLWcuI9Z1o6FDFJna5emddYQFPX+J1eaVJMHKpMIsYtEmf92UW+Yt1j8LBxY2fx3nOuO16bSXV7LecHnjeomSyLfOexccb1FDWX8Gzyi+dMOe02dvPfrE24W7uyPnTtyB5qlEgiJ4QQ44TRZOLVzZlY6zTctC5aplT2Q2el4WeXzeCCBYHsTi7j2U3Hae88e63D4cwqSmvauGxp6KDXt7X3dJBSk0aCdxw6jW40Qh8Ubzsvnlj8O+b7zOn3vHDnUAqbi+kx4/5uXxV+gwpY18taELVKzWK/+ZxsKqS8tdJs9xSQ9N20ytmSyIlRNPu76ZXJNeN3T7nshjwau5p6nRWhVqlZEbCUguZi8pu+3yBcURS+KdqNm7XrOeuJ+xPvOZO7Zv6UqvYa/p70wpk96wA2F2ynpqOO66KvQG/B3wf9kUROCCHGiS2HiimoaOHGtVE42Y3PXxrjiVql4srzwvnpBdFkFTXwp7eTqGk8NT3GYDTx6b58Aj3tmRvjOeg2E6uS6TEZWGShaZU/pFYN/Cs6zDmEHpOB4uZSs9yzur2WQ5VJLPZbgIu1c6/nLPBOQKvW8m25jMqZ07HqVEKdgvvsdyHMwdfeGx87rzPrMcejQxWJ2GptiHWL6fX4fO852Ght2Fny/QbhJ5sKKWguZlXgsiFvDTDNLYr74m6npaeVvyY9T2VbNUXNJewo3sti33lEuYaP6HlGkyRyQggxDhRXtfD5twXMi/FkbvTgEw8By2b58v+umkVjSxdPvJlIXlkT+1LLqWnsZMPyUNRDGNk8UHEUP3sfAh38RzFi8wlzCgYw2/TKrwp3oFGpWRu0os9z7HV2zPaI5XDFsQGLBIjBqWyrorytkjlS5ESMgTmes8hvKqSxa3CVf8dSe08HqTXpJHjN7nVtMJwqWLLEdz4p1WnUfbdB+DfFe7CzsmWhT8Kw7hvmHMzPZ9+F0WTk78de4I3M93DSO3JZ+EXDfpaxIImcEEJYmMFo4uUvT2BnY8UNa6IsHc6EFBPsyu9umoONTstT7yTz8d58IvydiB3C/nulLeWUtJSx0GfuhJnW6qCzx8vW0ywFTyrbqjlSeYxlfotw0jv2e+4SvwV0GjtJqhr/RRMmgqTq46hQWXRdppg6zkyvrB5/RYuSqlPpMRlYMMC08uX+i77bIHw/lW1VpNVmstxv0YimxAc4+PLAnJ9hpbaiqr2Ga6Iu63e/uvFAEjkhhLCwLw8UUlrTyk/XRWNv0/snkGJgPm52/O6mOYT4ONDWaeDy5WFDSshSatJQoSLBK270ghwF4c4hnGwsxKScW/BlKLYUbMdKY8XqoPMGPDfMKRgfOy8pemIGiqJwrCqVcOeQARNoIczB284TP3ufcVm98nBFIr523gPOinCxdibecyYHyo/wZcF2rNRalvkvGvH9vWw9eDDhXu6etZFY92kjbm+0SSInhBAW1NFlYNvREubFeBIXMbZ7lk1GDrY6Hrx2Nk/ePp/IAOchXXu8NpNQp2AcdPajE9woCXcOodPYSdkIio+09bRzrPo4S/0WDOr5VSoVS/wWUNRSYrb1eVNVeVslle3VsnecGFPxnjPJbyqiobPR0qGcUdlWRUFzMfN/sHdcf1YGLKXT2EVy9XEW+sw127/dTnpHprtFm6Wt0SaJnBBCWNCB9Eq6uo2snRdo6VAmDa1GjY+b3ZCuqetooKy1gpke4/8T2B8LcwoBRrZOLr+pEAWlz+ICvZnnFY+V2kqKnozQse+mVcZ5Dr7SnhAjdXqbi6RxNCp3qCIJtUrNXK/4gU8GghwDCHMKQYWKlQHLRjm68UkSOSGEsBBFUdiVXEawtwMhPjKlypLS6jIBJsRUmh9zs3HBRe9MXmP+sNs42ViIRqUhyHHwHyjYWtmQ4BXH0aoUOgydw773VHZ6WmWkSxiOOgdLhyOmEE9bD0Icg/gifxv7yg6hKIpF4zGajBypTGK6WxRO+sH/LFwffTm3xd6Ih+3g10NPJlpLByCEEFNVTkkj5bVt3HLhxJjCMZml1WTiZeuBl62HpUMZlnDnELIaclEUZViFWvIaCwh08EfXR5W4viz1W8DBiqMcrUxmmf/CId93qittraC6o5ZVgVNzNEFY1p0zb+aNzPd4L/tj8hrzuTZqA9Za61G7X15jAV+Xn8TUrcJaY41eq8dGo0ev1VPTXktTdwsLvIdWddLLzhMvu6lb6VkSOSGEsJCdx8qws9YyL8bL0qFMaR2GDnIb81kRsMTSoQxbmHMIR6uSqemoxXOIyWi3sYfiltJhPX+ggz9eth4cr82QRG4YjlWnolapifOQapVi7Dno7Ll71ka+LtrFl/lfU9xSym0zbsTP3sfs98prLOAfKS9hVIx9jv7ZW9kxw33w07uFJHJCCGERja1dHMup4fwEf/RWQ9u8VJhXZl0ORsU4IadVnhbufGqdXF5j4ZATuaLmEoyK8UwbQ6FSqYh2jeRA+RF6jD197vskzqUoCklVqUS5hGOvG9qaTiHMRa1Ssy54FaFOwbyW8Q5PJ/6DKyMvYZHPPLNtw1LeWsm/j7+Om7ULf1rza1obe+g0dtFp6KLT0Pndnztxt3FDq5bUZChkjZwQQljA3pRyjCaF82b7WTqUKS+tNhM7K1tCnYIsHcqwedt6Ym9lN6x1cie/24MuZJjPH+MaQY+ph/ymomFdP1UVt5RS11kv1SrFuBDpEsb/zPsFoU7BvJP1EW9kvk+noWvE7dZ3NvCv1FfQqa24Z9ZtOOjt0aq12FvZ4W7jir+DL+HOIcxwj8F7Ck+RHC5J5IQQYowZjCZ2p5QxI8QVLxdbS4czpRlNRtLrspjhFoNaNXF/JapUKsKcgodVufJkYyE+dl7YWw1vVCjCORS1Sk1WQ+6wrp+qkqpT0ag0xHlMt3QoQgDgqHPg3rjbuChkNYlVybye+e6I2mvraedfKa/QZezinrhbcbNxMVOk4rSJ+1tLCCEmqJTcWhpbu1kRL6NxlnayqZAOQwczJ/C0ytPCnEOo7aynsatp0NeYFBP5TUWEOQUP+77WWmtCHIPIqs8ZdhtTzalqlceJdo3A1ko+zBHjh1ql5sKQ1awNWkF67Ylh7zPXbezmhdTXqO2s587Ym0dl3Z2QNXJCCDHmdiWX4eaoZ1aYbABuaWm1mWhVGqJdIy0dyoh9v06ugASvuEFdU9ZaSaexk7BhrI/7oRjXCDYXbKe1u03We3HqTezOkm8pbC7CwcoBR70DTjoHHHUOOOodaelupaGrkZ+ErrV0qEL0aqHvXL4q2smhiiQuCFk1pGuNJiOvZvyXwuZibp1xAxEuYaMUpZBETgghxlB5bRsnihq4fHkoarV5FpKL4VEUheO1mUS6hmOt1Vs6nBHzt/dFr9FxcgiJ3OmpmKc3FR+uaNcIviz4muyGXOYM8t4ThdFkpMvYja2VzYDnmhQTiVUpfHZyK41dTXjZelJkKKWluxWFsyv1aVWaCbkBvZga3G3ciHQJ52DFUdYGrxj01HNFUXgv+xPSak9wdeSlzPaUiqyjSRI5IYQYQ7uSy9BqVCyd6WvpUKa8yvZqajvqOH+S7OGlUWsIcQwibwjr5E42FeCsd8LV2nlE9w508MdGa01W/eRK5Oo6GvhX6stUt9cS6RLGHK9ZxHnEYtfLdMi8xgI+zv2SopYSAh38+Om0a4lwCQVOJYOtPe00dzfT3N1CU1cLLtZO2GgHTg6FsJSFPgm8kfkeeY35RLqED+qar4t2caDiCOuCV7HMf9EoRygkkRNCiDHS2W3gQHoFCdGeONrpLB3OlJdWkwnADLfJs29RuHMImwu209bT3muy8UOKonCysZAIl9ARlxnXqDVEuoRzon74m5KPN+Wtlfwr9VShhhUBS0irzeSdrI94L/sTYlwjmeM5i5ke02nraePTvC0k16ThrHfippirmes9+6wRDI1ag5PeASe9gwWfSIihifOI5QPtpxwoTxxUItfc3cLWwh3EecSyPmTNGEQoJJETQogxciizio4uIytn+1s6FAEcr80k0MEPlxGORo0n4c4hKCicbCxg5gDVEOs662nqbh5RoZMfinaJILUmneqOWryGuJfdeJPfVMQLqa9ipdbyQPzP8LP3YUP4ekpay0iqSiWpKpWMuiy02VpQFNQqNReFrGZV4HL0GvmQRkwOOo0VCV6zOVRxlPaeSwacXry9aDcGk4GLw9ZNig9zJgJJ5IQQYgwoisLOpDICPO0J83O0dDhTXnN3C4XNxVwYcr6lQzGrYMdAbLU2HKk8NmAid7KxEGDEhU5Oi3aNACCrPndCJ3IZdVm8nPYWTnpH7o27HXcbV+DUFg+BDv4EOvhzSdgFFDaXcKwqFRMKa4LOw1nvZOHIhTC/hT4J7Cs7SFJ1Ckv9FvZ5XlNXC/vKDjHXe/aE/vmfaGT7ASGEGAN5ZU2U1rSyIt5PPqkcB9Jrs1BQiHWfXHt4WWmsWOw7n5SadOo6Gvo9N6+xAButNT52Xma5t4eNG27WrmTVT9z95I5UHuPfx1/Hy9aDX86550wS92NqlZpQpyCuiLyYqyIvkSROTFqBDv742nlzsDyx3/O+Kd6NUTFyQfDQKlyKkZFETgghxsCuY2XY6DUsnOZt6VCGTVEUFEUZ+EQLMykmylsr+401rTYTF70z/pNwb6Nl/gtRqVTsLTvQ73knmwoJdQo220boKpWKaNcIchpOYjQZzdLmWNpV8i1vZL5HuFMIP4+/CwedvaVDEsLiVCoVi3znUdRSQllrRa/nNHU1s6/sIHO9ZuMpo3FjShI5IYQYZa0dPSRmV7Noug96ncbS4QyLoii8m/0RD+17jE/yNlPXUW/pkPq0q+RbnjzyN15Kf4umrpZzjncbezhRn0Os+7RJOTrqau1CnMcM9pcfodPQ1es5Ld2tVLVXEz7CbQd+LNo1gk5jJ0UtJWZtd7R9U7yHTbmfE+cxg7tnbcRGa23pkIQYN+Z6zUaj0nCw4mivx7cX7caomLggeHJNVZ8IJJETQohRdjizCoNRYcnMiTv6s7/8MPvLj+Bm48LOkn384eBfePH4G2R9V6VwvDApJvaUHsBF70xGXRZPHH6GwxVJZ8WY3ZBLj6mHme6Tdw+vFQFL6DB0cKQyqdfj+U2FAIQ6B5v1vlEu4ahQccIC0ys7DJ28nvEemwu2D+m6TkMnWwu+YYZbNLfOuAErjdUoRSjExGSvs2Omx3SOVB7DYDKcdayxq4l95YeY5x2Ph62bhSKcuqTYiRBCjLJvj1cQ6GlPkPfELD1e1FzChzmfMc01ip/NuuW7aTSH2F9+mOO1GXjbebHcbxHzvOMtvrH2ifoc6jrruWX6dQTY+/J21ibePPE+iVUpXBu9AVdrF9JqM7HW6An/bo+vySjEMYgghwB2l+5nid+Cc6ZPnmwsRKvWEuQYYNb72lnZEujgT1Z9DheFrDZr2/2p7ajn38dfo6KtCo1Kw2LfeYNet3a48hidxi7WBZ9vtmmmQkw2C33mklx9nOO1mcR7zjzz+tdFuzEpJlkbZyHyL5YQQoyi4qoWiqpaJuxoXGtPGy+nv42DzoGbp1+DWqXGxdqZi8PW8cSi33JjzFXo1Frez/n/7N13fJX13fj/13Wdnb13GCGLTdgbGbIR3LirrVqt1Xo7atv7vvutv/u+O62t1aqtbd1b2UNBBVmyNwQIkBCy9zz7un5/RKM0CSThZJH38/HgATnXel9vzjk57/NZy/i/XX/E7nF0abxb8nYQaApgROQQov2jeHTkD7kxdTFZVWf5351/ZEveDg6XHmdgeBom9cr9LlNRFKYnTqaovoTj5SebbM+qOkvfwIQOyUF6WArZ1bnYPXafn7s5WZVn+f2ev1DlrOa29BsaW2Vb45t9+wYm0j+4TwdHKkTPNTAshRBL8AXdKyudVWzL38n4mFFE2KQ1ritIISeEEB1o6+ECjAaF8YN73iQnmq7x2tF3qXZWc+/QOwgw+V+w3WQwMT52NE+OfpgHh99DuaOSdWc3dlG0UGYv50hpJpPixmL8ukBRFZWrEibxi7H/Qd+gRN49sYxqV80V3a3yGxlRQwk2B/FF7tYLHnd6XeTW5Pls2YF/lx6WgqZrnKw40yHn/64d+bt5bv/f8DPZeGL0Q0yMG8vwyCFsyfuqxfGB33WiPIui+mKmJUzs8FiF6MlURWV87GiOl52kwlEJwCfZX6DpGnOlNa7LSCEnhBAdxO3R+OpoESNSIgmw9bxxN+uyP+NY+QluSF180S54iqIwODydCbFj+OL8Vgrqijoxym9tzd8JwKT4cU22RdjC+PGIe7kt/QaGhA9kaMTAzg6v0xlVI1MTJnC8/OQF/yc51efQdM1nC4H/u/7BfTEbzB26DIGma3yctZo3Mz8gJSSJJ0Y91Dhb3sw+U7F77C1OzPBdm85vI9AUwMjo4R0WqxBXivExo9HR2Vm4lwpHJdvzdzIhdjThLSzTITqeFHJCCNFBDmaVUmt3M6ULulVWOCrJrjjf7uOPlp1g3dmNjIsZxeS4poVRc64ZMBeLwcL7J1d0+gQobs3D9vxdDIkYSJg1tNl9vplG+4Hhd2Mz2jo1vq4yKW4cJtXIpu+0ymVVnkVBIamDCjmTaiQlJInMiqZdOn3B7nHw8qHX+Ozcl0xLmMiDw+/Bz+TXuD0puC9JwX35InfrRZdBKKkv42hZJpPjx13R3WyF8JVIv3BSQpLYkb+b9Tmfo6Ezp6+0xnUlKeSEEKKDbD1cQGighcH9OvfbSofHwbP7XuLJT/+XFw/+i9ya/DYdX2av4LWj7xDrH83StGtbPUV/oDmAa5LmcLIii33Fh9oTersdLD5MrbuOqfETOvW63V2gOYAx0SPZWbiPWncd0DDRSVxADH6mjitm08NSKK4vpdxx8UXJW6vMXsH2/F3888hb/HLHbzhWfoKbU5dwU+oSDGrTJT1mJk6lzFHOwdKjLZ7zy7ztKIrC5PjxPolRiN5gYtxYSh3lbM37igmxYwi3Nf/Fmegc8hWUEEJ0gIoaJ4fPlDF/fF9UtXPXKvvo1CrKHRXMS5nOprNf8ZvdfyIjahgL+19NjH/0RY91ax5eOfIGXl3j3qF3YDaY23TtyfHj2Za/i4+zVjM4PL3TZrH8Mm8HEbZw0sNSOuV6Pcn0xMlsL9jF9rxdzOwzlbPVOYyLGdWh10wPbfh/yCw/xcS4sW0+3uFxkFl+isyKLE6Un6LYXgpAsDmQweHpTIobR/JFxvgNixxMhC2cz859SUbk0CZfRjg8TnYU7CYjcmirZ7cUQsCIyCG8Z7Di1tzM7Tejq8Pp9aSQE0KIDrD9SAG6zmXPVvlNF8XWtoodLDnK9oLdzO47nbtH3sSMmKv4PPdLPs/dwoHiw4yJyWBB/6uJsIWj6Rol9jLyagvIry0kv66Q3Jo8yh0V3Df0zsYxR22hKio3py3hmb1/5ZOcz1k8YF6bz9FWebUFnK7K5trkBTJ9fDPiAmJIC01mc952UkKTcHpdHTY+7hux/tEEm4PaVci5vG5+u/s5iu2lWAxmUkIGMDVhIulhKcT4RbXqtaAqKjMSp/D+yeWcqcphwL+tl7ercB92j4OrEie1KTYhejuzwcz1KQvxaJ4Wu7GLziOFnBBC+Jiu62w9VEBqQjDRoX6XPqAZmq6xv/gwa85+ir/JjweG3X3BOKDm1LhqeTvzQxIC4hrX8PIz2ViYNIerEiaz4dwmNp/fzp6iA8T6R1NcX4pbcwOgoBDpF06fwAQWJc1heOSQdsUNkBTcj3Exo/js3JeMjx1NdDsKwrbYkvcVRtXI+NjRHXqdnmx64mReOvQqH2etAeiwGSu/oSgK6WEpHCk7jqZrbSqwP835nGJ7KXcPvpWMyKHNdp1sjfGxo1l95hM+O7f5gkJO13U2520nMTCe/kF923VuIXqz9rSyi44hhZwQQvjYqfNVFFXYWTChX7uOzyw/xYrTazlXk0e0XxTnqs/z3P6/8VDGvU2WAPiGruu8nfkRDo+DuzLub5x+/xsBZn+uTV7AjMQpbMjZRGF9MWmhycQFxBLvH0OMfzRmg+9m1lySPJ+DJUf54OQKfjT8+61uUWwru8fBrsK9jIoa3mJuBAwOTyfSFs6ZqmzCrKGEWkM6/JrpYSnsLNzL+Zp8+gQltOqY4vpSNuRsYnT0CEZHj7is61sMZqbGT+CTnC8ori9pbGE+UZFFYV0Rtw+8qcOel0II0RmkkBNCCB/beqgAi9nA6PS2tUTlVOey4vQ6TlRkEWYN5c6BNzMmJoPj5af42+HX+PO+l3k44z4CzQFNjv2qYA+HSo9yXfJC4gJaXrMu2BLEDanXtPme2irIHMjCpNl8eGolB0uPMuIiLXxezYtLc+PyunFrrq//duPyurAarSQGxrd47O7CfTi9LqYmyCQnF6MqKlclTuaDkysYENyxrXHfSPt6nNzh0mOtKuR0XeeDUyswqkauS17okximJkxi47nNfJG7lZvTrgUalhwIMPkzOkqWHBBC9GxSyAkhhA85XB52ZxYzZmAUVnPr3mJL6stYcXot+0sOE2Dy5/qURUyJn9A4Jfrg8DQeGHY3Lx16lT/te4mHM+4j2BLUeHypvZwPTq0gJSSJ6YmTO+S+2mNq/AS25+/io1OrGBSW2jhxSpWzhqzK05ysPMOpijMU1Rdf9DxDwgdyXcrCJl00dV3ny7wdJAbG0zew5XXuRIPxMaP5qmAPo6KHdcr1gi2BDAlP55OcL0gNHUBK6ICL7n+w9CjHyk5wfcqiC57flxvDmJiR7CjYw4Kk2Tg8To6UHmd23+mYfNgCLYQQXUEKOSGE8KHdmcU43d5Wrx1X66rj2X1/xe51Mq/fLGb2mYrNaG2yX3pYCj8afg9/PfSvxmIu1BqCpmu8fuw9FFTuGHhzt5rsw6AauCl1CX/a/xJvZ36ExWDmVOXZxsLNarAwIKQ/I6OGYTVaMKtmzAYTZoMZs2rCpJo4V3Oe9dmf8T87n+GqhEnM6zercdr801XZFNQVcVv6DdJFrhWsRgtPjXmkU69516Cl/GHvX/nb4dd5fPRDLY6XdHpdfHhyJXH+MUyLn+jTGGYkTmFHwW62nP8Ku9eOoihMkSUHhBBXACnkhBCiDS41i+TWQwVEh/mRHH/pKc11XeetzA+pc9fz+OgfkxgYd9H9U0IH8OMRP+CFA//k2X0v8UjGfewrPsTpqrPcOfDmbrmeT0poEmOiM9hdtB+rwUJySH8mxI4mNXQACQFxl5zIIi0smXGxo1h95hO+yN3KrsJ9LEyazcTYsXx5fjs2o/Wyx1KJjuP39UQ9f9j7PC8d/BePjf5Rs2MZP8n+nApnJd8b/EC7JzdpSVxADIPC0th8fhse3cvwyCGdMkZQCCE6mhRyQgjRSpqm8/Sru6mud5HeN5T0PqEM7BtKZEhDC1FheT2nzldxw1UDWtVCtC1/Z+O4tksVcd9ICu7Hwxn38pcDr/DHfS9S66plROQQxsaMvKx760i3pt/A1X2vIsYvql0f0oPMgdyafgNT4ify0amVvHtiGV+e30FRfQlTEya0ea070bki/cK5b+hdPLf/Zf5++HUeGnFvY7dhgKK6Yjae28y4mFEXXRvucszsM5W/HPg7AFclyJIDQogrgxRyQgjRSntPlnCuuJb0PiEcO1vOV0eLAAgPspLeN4R6hwdFgQmDW55s5BtFdcV8dGoV6aEpbR7X1jcokUcy7ucvB/6GzWTjlrTru3XXQrPBRHzA5a2nB5AYGMcjGfdzoOQIy7JWo6MzJU66yPUEA0L6ccfAm/jXsXd4J/Mj7vh6xkhd13n/5ArMBhNLkud32PXTQpPp8/WkOR29hp4QQnQWKeSEEKIVdF1n7Vc5RIfaeHxpBooC+WX1ZOZUkJlTwYFTpdQ5PIxIjiA00HLRc3k0D68eeweTauKOQTe1a1xbYmAcvxj7H2i6RoC590y7rygKGVFDGRKeTqWzmki/8K4OSbTS6JgMiu2lrDm7gSi/COb2m8lX5/eRWXGKG1MXE2QO7LBrK4rCj0fc2/hvIYS4ElxWIbdq1SpefPFFPB4Pd911F7fddtsF259//nk++ugjgoIaZp+66aabmuwjhBA9wfGcCnIKa7hzbhqq2vBBMD7Cn/gIf2aOSkDTdfJL6wi7RBEHsObsBs7V5HHv0DsJsVx6LF1LfDWzX09kMpikiOuB5vWbRXF9GavOfEKQOZB1ORtJCIjrlJZVP5Nfh19DCCE6U7sLuaKiIp599lk+/vhjzGYzS5cuZdy4cSQnJzfuc+TIEf74xz+SkZHhk2CFEKKrrPsqhyB/M5OGNN9tUlUUEiKbru/2705WnGZDziYmxY296NpqQlyJFEXhtoE3UO4o563MDwG4e9RtPp/gRAgheoN2z1O9fft2xo8fT0hICH5+fsyZM4f169dfsM+RI0d4+eWXWbRoEU8//TROp/OyAxZCiM6WU1jD0ewKrh6dgMnY/g+c9e56Xjv2LpF+4Vyf0vGLcgvRHZlUI/cNvYv4gFjmp0wnKbhvV4ckhBA9UrsLueLiYiIjv10PJioqiqKiosaf6+rqGDhwIE888QTLli2jurqav/71r5cXrRBCdIF1O3Owmg1Mz4hv9zl0XeftEx9T7arhe4NuwSIzLYpeLMDsz8/G/ITvjbypq0MRQogeq91dKzVNu2DAsK7rF/zs7+/P3//+98af77nnHn7+85/z6KOPtvoa4eGX7qYk2iYysuMGk/dGkk/f66ycns2vIjLERoDfxQuqgtI69mQWs2RaMn0Tw9p9vU1nd7C/+BC3DF3M6AGD2n2etpLnqG9JPn1Pcupbkk/fk5z6luTTd9pdyMXExLBnz57Gn0tKSoiKimr8OT8/n+3bt3PDDTcADYWe0di2y5WV1aJpentDFP8mMjKQkpKarg7jiiH59L3OyumOo4W8suoYsRH+/PTWDAIvUsy988kJVFVh0uDodsdW5azhX/veZ0BwfyZGTOi05408R31L8ul7klPfknz6nuTUtySfbaOqykUbttrdtXLixIns2LGD8vJy7HY7n376KVOnTm3cbrVa+f3vf09ubi66rvPWW29x9dVXt/dyQgjhE18dK+SV1cfoFxtIcYWdP75/ELvT0+y+1XUuth4uYMLgmEsuKXAxy7LW4Pa6uW3gDe1aakAIIYQQ4t+1+xNFdHQ0jz76KHfeeSdLlixh4cKFDBs2jHvvvZfDhw8TFhbG008/zQMPPMDcuXPRdZ27777bl7ELIUSb7DpexN9XHSMtMYQnbxnJg9cO4XxxLX/+4CBOt7fJ/hv3nsfj0Zg7rk+7r3myIovdRfuY1fcqov0iL32AEEIIIUQrXNY6cosWLWLRokUXPPbdcXFz5sxhzpw5l3MJIYTwiT2Zxfxt5TFS4oN55IbhWMwGRiRHcO+iQby84igvLDvMw9cPw2ho+H7L4fLwxb7zZKRGEhvevgW3PZqHd08sJ9waxpy+M3x5O0IIIYTo5aSPjxDiirf3RAkvrzxKUnwQP7mpoYj7xtiB0dw1L50jZ8p5eeVRvJoGwJcH8qlzeJg3vv2tcZ+f20JRfTE3pS7GbDBd9n0IIYQQQnxDCjkhxBVt/8kSXlpxhH6xgTx643Cs5qYdEaYOj2PpzBT2nijh1XWZuD0an+zOJS0xhAFxwe26bpm9grXZGxkeOYQhEQMv9zaEEEIIIS5wWV0rhRCiOztwqpS/Lj9C35hA/uOmEdgsLb/lzR6TiMPpYfnWs+SV1FFR4+SuuentvvYHp1agADekLLrkvkIIIYQQbSWFnBDiilRUXs+LK47QJzrgkkXcNxZN6ofd5eGTXbkkRAYwNKl968YdKjnK4dJjLBkwnzBraLvOIYQQQghxMVLICSGuOLqu8/onJzAaFB66bhh+1ta91SmKwk3Tk4kItpEcH4yiKG2+tsvr4oNTK4nxj2ZG4pQ2Hy+EEEII0RoyRk60m8eroemyYLvofrYfKeR4TgU3TBvQ5vXfFEVh5qgE+sYEtuva67M/p9xRwdLUazGohksfIIQQQgjRDlLIiXZxezT+9/W9/P7t/Xi8WleHI0SjmnoX732exYD4IKZlxHfqtQvrith4bjPjYkaREprUqdcWQgghRO8ihZxol1Xbs8kpquFEbiUfbjrd1eEI0ej9z7OwOz3cNTcdtR1dI9tL13XeP7kCs8HMtckLOu26QgghhOidZIycaLOcwhrW7shh0pAYrBYjn+7OJTk+mNHpUV0dmujljmeXs+1IIQsm9CUhMqBTr320LJMTFVncmLKYQHPnXlsIIYQQvY8UcqJNPF6Nf609TqCfiZtnpmA1GzhbUM0/1x4nMSqA6DC/rg5R9FJuj5fXPzlBVIiNRRP7deq1NV1jxel1RNrCmRw/rlOvLYQQQojeSbpWijZZt/Mc54pruWNOGgE2E0aDygOLh2BQFV5YdgSn29vVIYpeatX2HIoq7NwxNw2zqXMnGdlZsJf8ukKuGTAPoyrfjwkhhBCi40khJ1otr6SWVdvOMnZgFCNTIxsfDw+2ct81g8krqeWtT092YYSit8orrWPdVzlMGBzN4H7tW/utvVxeF6vPfkrfoEQyIod26rWFEEII0XtJISdaxatp/HPtcaxmI7dendpk+9CkcBZO7MfWwwVsOZjfBRGK3krTdV5bn4nVbODmmSkXbCuoK8LldXXo9b/I3Uqls4prByxo17pzQgghhBDtIYWcaJUNu89ztqCG265OJcjP3Ow+iyf3Z2DfUN7ccJJzRTWdHKHorb48kE/W+SpumpF8wXMztyaf/9v1LK8ceRO9g9Y7rHXV8WnOJoZGDJLlBoQQQgjRqaSQE5dUWF7Psi1nyEiJYOzAlmemVFWF+68ZjL/VyF+XH6He4enEKEVv4nB52HmsiBc+PszbG0+S3ieEyUNjG7drusY7Jz4CGmaT3FN0oEPiWJ/9GU6vk8UD5nXI+YUQQgghWiKj8sVFabrOv9Yex2RQuWNO2iW7jgX5m/nh4iH87u39vLruOA8sGSLdzYRP2J0eDp4uZU9mCYfPlOH2aAT7m5k6PI6FE/td8Dzblr+TnOpc7hh4E1vyvuLDUytJD0tp1bIAHs3DOyc+JsIazuy+V2FQm584pdRexpd5O5gYN4ZY/2if3acQQgghRGtIISea0HWdmno3xZV29p0s4dT5Kr6/YCAhAZZWHZ+aGML105L4YNNpvjyYz7QR8R0csbgS1TncZBfUcLagmtN5VRzLqWgo3gLMTB0Wx+j0SFISQlDVC78oqHbVsOL0OlJDBjAuZhR9AhP4ze4/8+Gpldw9+NaLXlPXdd49sYyvCvYAcKj0KHcNWkqMf9OW6JWn12NQVOb3v9p3Ny2EEEII0UpSyAkOZpWSea6CkkoHxRV2SqrsOF3fLiOQkRLBxCExbTrnnHF9OJpdzjsbT5GSEEJchL+vwxZXmPPFtWw7VsyRrBLOFlRTXGFv3BYdamPq8DjGpEeRnBCMepFW3o9PrcbtdbM07VoURSEuIIY5faezNnsjY6IzGBIxsMVjN57bzI6C3czrN5O4gFjezfyY3+z+E4sHzGdawkRUpaE3ek51LnuLDzKv30xCLMG+S4IQQgghRCtJIdeL2Z0e3vz0JDuOFmI0qESGWIkMsZHeJ4TIUBuRITaiQmzEhPm1uXukqij8YOEg/vsfu3h55VH+885RmIydu7aX6DnyS+v41au78Wo6oYEW+sUEMmVYLP1ig+gXE4i/1dSq82SWn2J30X7m9ZtJ9Hda0Wb3m8H+ksO8c+Jj/jPkMWxGa5NjD5YcZcXpdWREDWN+/6tRFZUBwf14O/NDPjy1kkOlx7hj4I2EWkJYlrWGAJM/s/pM81kOhBBCCCHaQgq5Xup0XhV/W3WU0ioH10zqx6JJ/TCovp37JiTAwvcXDOTPHx7ig02nuXVW02ULhAD4bO95FEXh5Z/NwNTOGSbdXjfvnVhGhC2c2X1nXLDNpBq5Lf0Gntn7V1aeXsfNaddesD23Jo9Xj75Nn8AE7hx4U2PLW7AliB8Ou5vtBbv46NQq/nfns4yLHcWpyjPclLoEazMFoRBCCCFEZ5BZK3sZTdNZte0sv35zH5oGT902kiVTknxexH1jeHIEs0YlsHHPeQ6dLu2Qa4ierd7hZtuRAsYPiiYu4tKTkbRkw7lNFNtLuTl1CWZD0xa8/sF9uSphEl/m7SCr8mzj41XOal469Cp+Jj/uH3YXZsOFy2soisKkuHH8fOyjxAfEsvn8NqJsEUyOG9fuWIUQQgghLpcUcr1IcUU9v3t7H8u2nGXMwCh+dc9YUhJCOvy6N04fQGJUAP9Yc5zKWmeHX0/0LFsOFeBya8wcldDucxTXl/BJzheMihrOoPC0FvdbmDSHcGsob2V+gNvrxuV18fKh16j32PnhsLsJtgS1eGyELZyfjLyf2wfexD1Dbm9xNkshhBBCiM4ghVwvsfdEMQ//4QvOFdfyg4UDuW/RIPysndOz1mQ0cP81g3G6vPxj9TG0DlqcWfQ8mqbz2d7zpCYE0zcmsF3n0HWd904sx6gYuS5l4UX3tRot3JJ2PcX1pazN3sjrx9/nXM157h50C4mBcZe8lqqoTIgd3ap9hRBCCCE6khRyvcCJcxW8uPwo8VEB/L97xjJxSGynr+0WF+HP0lkpHM2u4NNduZ16bdF9HTxdSmmVg1mjE9t9jr3FB8msOMWiAXNaNYPkwPBUxsWM4tOcL9hffIglyfMZFjm43dcXQgghhOgKMtnJFa6ixsmLK44SFWrj/7t/InU1ji6LZdrwOI6eKeejzadJ6xNC/9iWu7GJ3mHjnvOEBlrISI1o87FOr4tt+TtZf/Yz+gTGMzV+QquPvT5lEaershkUlsrMxKltvrYQQgghRFeTQu4K5vFqvLjiCA6XhyeWjsDPampzIffl+R1YDGbGxY667HgUReGueemc+ecuXlpxhP/+3phWTysvrjx5JbUcz6ng+mltm2yn1l3H5vPb2Zy7jTpPPSkhSdySdl3jTJOt4W/y45fjn2jTMUIIIYQQ3YkUclewD744Tdb5Ku6/ZjDxkW2fDTCvtoD3Ty5HR6fEXsqC/rMvu0tmgM3EA0uG8Nu39vHKqmP8+IZhF13cWVy5Ptt7HpNRZerw1o03q3BU8nnuFrbm78TldTE0YhCz+04nKbhvu64vRZwQQgghejIp5K5Qu44XsWFPLrNGJTBuUHS7zrHi9DqsRitDIwayLvszql21LE279rI/ACfHB7N0ZgpvbTjJ6u3ZXDOp/2WdT/Qcuq6jKAp1DjfbjxQyflA0gX7mZvf1al4K6orIqcnlVMUZ9hUfQkdndPQIru5zFXEBMZ0cvRBCCCFE9yGF3BUor7SOf63NJDk+mJtmJLfrHCcrTnO0LJMlA+Yzq880QizBfJrzBbXuOu4edAumZtbpaosZI+M5k1/Fii1n6RcTxLAB4Zd1PtG9eTUvn+V+yfrszzAqRoyaP/RTUeL7sym3nHBbKMHmIE7Z6zh0/iQ51bmcq8nDrbkB8DPamBQ3jll9phJuC+viuxFCCCGE6HpSyF1h7E4Pf112GItJ5YElQzAa2t56pus6y7PWEmIJZlrCJBRFYfGAeQSZA/nw1EpeOPgP7h92Fzajrdljc2pyOVB8hCi/CCbEjmm2O6aiKNw5N53c4jr+vuoo//29MUSGND2f6PmyKs/y7omPKagrYkj4QEIswWw/eRZLoJ0DFXvZXea+YH+TaiQxMJ7J8ePoG5hI36BEIm3hnT7TqhBCCCFEdyaF3BVE13X+tfY4heX1PLE0g9BAS7vOs6/4EDk1udw+8CbM32l5m544mQCTP68ff49n973Ej4Z/n2BLELquc742n71FB9lXfIgyRzkKCjo6ebUFXJ+yqNnumBaTgYeuG8LTr+7hhY8P8/M7RmE2ySLLV4padx0rstayvWA3oZYQfjjsewyNGMS+kyVsOBbCj64dwsjUSGrddZQ5yql0VJEcl4jNFSiLbQshhBBCXIIUcleQT3fnsudECTdeNYD0vqGNj7u9bjae28wQdzKJpn4XPYdX87LyzHri/GMYFzOyyfYxMRkEmPz525HXeWbvXxkVPZwDxYcptpeiKippocnM6zeTYZGDWZe9kS9yt1LtquHOQUsxqU2fblGhfty7aBB//vAQb3xygnsWDJSWlx5O13V2Fu5lWdYa6j12ru5zFfP6z8JiaBgLt3FPLuFBFkakRKAoCoHmAALNARAEkaGBlJTUdPEdCCGEEEJ0f1LIXSGKyuv5cNNpMlIimDuuT+PjxfUl/PPIW+TW5rM2eyPfG7SUUdEjWjzP1vydlNrLeGDY3S1OajIwPJVHMu7jrwf/yYacTaSGDmBWn2kMjxxCgNm/cb/rkxcRYglmWdYaat313Df0TmxGa5PzDU+O4JpJ/Vi5LZuk+GCmZ8S3PxGiS1U4Knnt2LucqjxDUnBflqZdR3xAbOP23OJaMs9VcuNVA9q05IAQQgghhLiQFHJXiPc+z8JoVLljTlpji9buwv28c+IjDIqBewbfyo7iXfzr6Dvous7omIwm53B4HKw9u4GUkCQGh6df9Hr9gvrw9ISf4dbcDa0pzVAUhVl9phFoCuDNzA/4076XeHD49wm2BDbZ95pJ/TlTUM3bG07SJzqAAXHB7ciC6EqnKk7zypE38Wgebk27nglxY5p8GfDZ3lzMRpUprVxyQAghhBBCNE++Er8CHD1bzoGsUhZO6EtIgAWX18Vbxz/g1WPvEB8Qx8/G/oRR0SP42ZQfkRzSn1ePvcuuwn1NzvPZuS+pddexJHl+q7o3Wo2WFou47xoXO4ofDrubYnspz+x9geL6kib7qKrCfYsGExpo4bV1ma27cdEt6LrO57lbeO7A3/E3+fPE6B8zKX5ckyLufHEtO44WMX5wDAE2WQheCCGEEOJySCHXzWi6zqptZzmTX92q/b2axrufnSIyxMrsMYnk1xby2z1/YUfBHub0ncFPMu4nzNowXs5qsvLA8HtICUni9WPvsbNgb+N5qpw1bMz9kozIofQL6tPS5dptcHgaP8m4H6fXyTN7/0pOdW6TfQJsJmaPSeR8SR35pXU+j0H4nsvr4tVj7/DRqVUMjRjEE6MfIsY/qsl+p/Or+O3b+wiwmVg4oX0LeAshhBBCiG9JIdfNbNydy7ItZ3n2/QMUV9ovuf+m/fnkldZx41XJ7Craw+/2/IU6Vx0/Gv59rhkwt8nsfxaDmQeG301aaDJvHH+fHfm7AViXvRGP5uGaAXM75L4A+gYl8tioBzGqRj44uaLZfUalRaEAezKLOywO4Rul9jL+sPcF9hYdZFHSXH4w5PZmx0Bm5lTwh3cP4Gc18rPbRhIhy0wIIYQQQlw2KeS6kfMltXy4+QzpfUIAeO7DQ9idnhb3r7W7Wb7lDMlJRrbbl/P2iY/oH9yXn419lIHhqS0eZzaYuX/Y90gPS+GtzA9ZfeYTtuXvZHLcOKL8In19WxeI8otkROQQ8uoK0XStyfbQQAvJCcHsPiGFXHd2tOwEv939HBWOSh4Yfg9z+81odnKcg1mlPPvBQcKDrDx12ygp4oQQQgghfEQKuW7C7dH4+6pj+FkM/HDxEB5cMoTCsnpeXnkUTdObPWbZ1tM4Q09SFLmenOrzLE27lh+P+EGzk4n8O7PBxP1D72JgWCrrsj/DqBqZ13+Wr2+rWTH+0bi8LiqdVc1uH50eRV5JHQVl0r2yu6lwVPL6sfd48eA/CbWG8NMxDzM4PK3ZfXcdL+L5jw8TF+HPT29t/7qGQgghhBCiKZm1sptYvuUMucW1PHz9MIL8zQT5h3Hb1Sm88elJPtx8mpumJ1+w/56ck2x3foApsYbBEUO4KXUxIZa2zfRoMpi4b+idfHBqJf2D+hBkvnQB6Asxfg1jqArqihvH733X6LQo3tl4ij2ZxSya1L9TYhIXZ/c4+DTnC77I3YIOzOoz7YK14f7dlwfzeW1dJskJwTxyw3D8rPJWI4QQQgjhS/Lpqhs4ca6C9TvPMXV4HCNSIhofnz4ygbzSOtbvPEd8hD+Thsbi8DhZfeYTvsjdhmoyc3vqrUxIGNHua5sMJm5Nv94Hd9F6sf7RABTVFTXbmtPYvTKzRAq5LubVvGzL38masxuoddcxJjqDRUlzCbc1LcChYQbLDbtzeffzLIb0D+NH1w3FYjI0u68QQgghhGg/KeS6mN3p4ZXVx4kMsbF0ZnKT7UtnplBQVs9r6zMx+dtZU/QBpfYyPMWJLB4wjwkJTY/p7gLM/gSY/Cmoa3kc3Ji0KN757BSF5fXEhPl1YnQCGlrgTpSfYuWZ9RTVl5ASksS1yQvoG5TY7P66rnPkbDkrtjbMuDoqLZL7Fg3GZJTe20IIIYQQHUEKuS729oaTlNc4+Nnto7Cam/53GA0qDywZwv97/xNey/oH/lYzfrlTMLsimT0qqQsi9o0Y/ygK61su5EalRfLOZ6fYnVnMoon9Oi+wXsbpdVFYV0R+XREFdYUU1BVRUFtEhbMSgGi/SO4fehdDIwY1u7agrusczW4o4E7nVRMeZOHOuWlMGRaLQZUiTgghhBCio0gh14X2ZBaz7Ughiyb2Izm+5fFtJ2uO4+qzDexWag+NxVFr4dGbUjAaeu4H5Rj/aPYVHUTX9WYLhLAgKwPigxrGyUkh1yF25O/m3RMf49G9ABhVI9F+kQwI6UesfwzxATEMCktrsoQFNBRwx3IqWLHlLFl5VYQFWbhzThqTh8X26OelEEIIIURPIYVcF6msdfL6JyfoFxPIokn9Wtzv83Nf8nHWGvoH92FqwmJeOnySYQPCGZoU3nnBdoAYvyjqPXaqXbUtzrI5Ji2Kdz/Poqi8nmjpXulTG89tZlnWGtJCk5maMJFY/2girGHNFm3fVWt3s/9kCVsOF5B1vorQQAt3zE5l8rA46UYphBBCCNGJpJDrArqu86+1mbjcXu5dNKjZFgxN1/j41Gq+OL+VEZFDuWvQUswGEzF3hxAZ0nTR5Z6mccKT+qIWC7nR6Q2F3J4TxSyY0K8To7ty6brOqjOf8EnO52REDeN7g5ZiVC/+NlDncLP/ZCm7M4s5ll2OV9OJDLFy29WpTB0uBZwQQgghRFeQQq4L5JXUcfhMGTdeNYDYcP8m211eN68ee4eDJUeYkTiFa5MXNC62nBgV0NnhdogY/2+XIEgNbX7ClrAgKwPigtideWUUcpquU2t343R5CQkwYzJ27myOmq7x/skVbMnbwaS4sSxNu67ZRby/iXXnsSJ2Hivi6NmG4i0i2MrsMYmMGRhF3+jAZrvECiGEEEKIziGFXBfYc6IYBZg4JOaCxx0eJwdLjvBF7hbO1xZwQ8o1TE+c3DVBdrBgcxBWg5XCi8xcCQ2tcu99nkVxRT1RoT2je2VFjZNN+/OorHVSVedq+FPrpLrOjaZ/u7h7oJ+J0EALYYFWwoIshAVZSYj0Z9iAiIucvX08mofXj73H3uKDXN3nKhYPmHfRQuyDL7L4ZFcuYUEWZo1OYOzAaPrFSPEmhBBCCNFdSCHXBfaeLCElIZjgAAuarnGy4jQ7C/dyoPgwLs1NhDWMe4fewfDIIV0daodRFKVh5sq6oovuNzqtoZDrSa1yy748w7bDBQT5mwkOMBPsbyExMuDrf5uxmA1U1rqoqHZQXuOktMrOydxK6p0eAJ7+/lgSIn3X8uryuvj7kTc4VnaCxQPmMbvv9Ivuv2FPLp/symXGyHhuuzpVijchhBBCiG5ICrlOVlheT15JHYumR7Li9Dp2Fe6j0lmF1WBlTEwG42JGkxTct1d8eI7xj+JY2YmL7hMebCUpLog9mSXdvpDTdI3DxafYXbeRmNHw4wk3N3YhbY3SKjtPvriD/SdLfFbI1brr+Nuh1zhTlcOtadczKX7cRfffe6KYdzeeIiMlgltnSREnhBBCCNFdSSHXyfaeKEax1bCpfiPaOY2BYalcl7yAoRGDMRtMXR1ep4rxi+Krgj3Uu+vxM7XcbXJ0WhTvf5FFcaWdqBBbJ0Z4abquc67mPHuKDrC36CBVrmqUUAMOk4nf7nmO29KuZ3RMRqvOFRFsIykuiP2nSlk0qf9lx5ZVeZZ/HX2bGlct9wy5jZFRwy6+//kq/rbqGElxQdx/zWBUVYo4IYQQQojuSgq5Trb7RBEBaUexGa08OebHhFlDuzqkLvPNzJWF9cUkBfdrcb/RaZG8/0UWezKLmT++bydFd3H17nreP7KJL8/sotheikExMDg8jawjAQR5E3joxnT+efRt/nXsHbKqsrk+eSGmVhTqGSkRfLT5DBU1TkIDLe2KTdM1NuRsYvXZTwmzhvL4qB/RJyjhoscUltfz3EeHCAu08PANwzCbOnciFiGEEEII0TYyb3gnKq20k284gMdcyS3p1/XqIg6+O3PlxcfJRYTY6B8byJ7Mi0+M0ll0XecfR97io6PrCLGGcFv6Dfxm8n8xNXgxZefCmDGiL6HWEH6ScT+z+kxjS94Ontn3V0rtZZc894iUSAAOZJW2K7ZqVw0vHPgHK8+sJyNyKE+NeeSSRVx1nYtn3z+AosCjNw0n0M/crmsLIYQQQojOI4VcJ/rs+FGMcWcYFjrsip7IpLXCrKGYVNMlZ66EhtkrswtrKKm0d0JkF7e9YBeZFae4Z+TNPJJxHxPjxuJn8mPT/jz8rUbGpDcUqAbVwLXJC7h/6F2U2sv5ze4/c7DkyEXPHRfuR1Sojf2nStocV2b5Kf5v17OcrjrLrWnXc/fgW7EZL77moNPl5c8fHqSq1sUjNwzvMTODCiGEEEL0dlLIdRK318326vUYvBbuGHJdV4fTLaiKSoxfZOsKubSG4mjPia5tlSt3VPDxqdWkhgzg6uQpjY9X1TrZd7KESUNjm3RLHBY5mKfGPEKkLYK/HX6dZVlr0L+zDMF3KYpCRkoEmTkV2L+exfJSNF1j9ZlPeP7AK/gZ/Xhi9I+ZFD/ukhOVeDWNl1YcIbuwhh8uHkJSXFCrrieEEEIIIbqeFHKd5KMT6/CYqhlhm3nRiT16m2j/qEt2rQSIDLHRLyaQPZltb6nyFV3XeTvzIzR0bht44wWLaW85VIBX07kqI77ZYyNsYfzHqAeZFDeWjec2c6y85dk6RyRH4PHqHDlb3qq4Vp/5lHXZnzEuZhQ/HfMw8QGxrbuXDac4eLqM269OZUSK79euE0IIIYQQHUcKuU5wpiqbLYVb8RQnMG/g6K4Op1uJ9Y+mwlmJw+O85L4ZKRFkF1RTXe/qhMia2lGwm+PlJ7l2wHwibGGNj2uazuYDeQzsG0pMWMtFukk1clPqEsKtoaw5s6HFVrnkhGACbKZWda/MLD/FpzlfMDF2DHcMugmLoXXj29bvOscX+/OYN64P00defAydEEIIIYTofqSQ62BOr4vXj72HweNPeG0GcRH+XR1StxLj19Blsqj+0l0mhySFowNHW9lS5UsVjko+OrWalJAkJsePv2DboTNllFU7md5Ca9x3GVUjc/vNJKcmlyNlx5vdx6CqDB8QzqGsMjxercVzVbtqePXYO0T7RXJj6uJW38uu40V88MVpxg6M4vqrBrT6OCGEEEII0X1IIdfBVpxeR4m9DHvWIEanXvqDfm8T880SBK0YJ9c3JpBAPxOHz1x69kdf0nWdtzI/RNO93P5vXSoBNu3PIzjA3OruieNiRhFhDWPN2ZZb5UakRFLv9HDqfFWz2zVd4/Vj7+HwOPj+kNsxt7Il7tT5Sl5ZfZyUhGC+v2Agqiz4LYQQQgjRI0kh14FOVmSx+fw2Uqwj8FaHMzotsqtD6nYibeGoikphK1rkVEVhSP8wjpwpR2uhAGpOXmkde0+0f2zdjoI9HC8/yeLk+UTYwi/YVlpp5/DpMqYOi8NoaN3LyaAamNt/Frk1eRwqPdbsPkP6h2Eyqi12r9x4bjPHy09yQ8o1xAXEtOq6heX1PPfhIcKDrfz4+mGYjLJWnBBCCCFETyWFXAepddfxxvEPiLJF4DmfSmSIlcSogK4Oq9sxqAaiWjlzJTR0r6y1u8kprGn1Nd7ecJK/LjvM6fzmW7cupqFL5SpSQpKYGj+hyfbNB/NBgWkj4tp03rHRGUTawllz9lM0vWn3SYvZwKC+oRw4Vdqk1e5MVQ6rznzCyKhhTIob16rrVdc3rBWnqgqP3jiMANulFycXQgghhBDdV48r5OweBytPr2fd2c/Ylr+Tw6XHyK4+R7mjArfWuunaO5pX8/LK4TeodlZzc/INnMiuYVRa1CWng++tYvyiKGzFzJUAg/uHoUCru1dW1TrJPFeBDry27sRFx5z9u8ZZKnUvt6U37VLp9mhsOZjP8AERhAVdfL22f2dQDczvfzV5tQUcKjna7D4ZqZGUVjk4X1LX+Fi9u55/HnmLUEsIt6Zf36rnlNPt5bkPD1FZ6+Lh64fJWnFCCCGEEFcAY1cH0FZf5G7hk5zPW9weYPJnfv+rmRI/HlVR0TSdE7mV7DxWyIlzlTx47dBWt4zpus5XhXtZdXods/pMY0afqU22HzlbTt/oQIL8zY2PvXdyOacqz3DnwJspL7Lh1XRGSbfKFsX6R3Gw5AhuzYNJvfhTMsjPTL/YQI6cKeeaSf0vee49J0rQdbh2ahLLvjzDxj3nmTuuT6vi2lW4j2PlJ7gxZTGRfuFNtn91uIDqejfTR7Zv7OPo6BGsz/6MNWc3MCxycJNCcfiAcBRg/6kSEqMCGsfqVbmqeWzUg9iMthbP7dU0aurdVNe5WLH1LGfzq3nw2qEMiA9uV6xCCCGEEKJ76VGFnMvrYvP57QwJT+cHQ++k1lVLtaum8U+Nq5YTFad5/+RyduYdJLpuPIeO11NR48RiMuDxamw+kMfts9Muea1adx3vZH7MgZLDBJoD+ChrNfUeBwv6X93YCnIsu4Jn3z+Iv9XI0pkpTBwSw6bz29iWv5PZfaczLnYUf9l+iNBAC/1jZbHllsT4RaGjU1xf0qo10IYmhbNqeza1dvcluwjuPFZEQqQ/Cyf05Wx+Ncu3nmF0WiQRIS0XQdDQqro2eyOJgfFMTWjapRJg3Y5sIoKtDO4f1uz2S1EVlfn9ZvGvY++wv/gwo6KHX7A9OMBCUlwQ+0+Vcs2k/mzJ28GBkiNcm7yAfkHfFqPFFfWs2pZNeY2T6joXVXUu6uxuvtshc+nMFPkyQQghhBDiCtKjCrmvCvZS665jVp+rMKlGQq0hhFpDGrd7vBqe/CRySwLI9h4mW88lNnEsNyWPZ0RKJP9Yc5w9mcXcMisFg9pyr9LjZSd54/h71LrrWTJgPtMTJ/PuiWWsy96I3WPn+pRFqIrKqm1nCQkwExFs4x9rjvPFqf0UhmxmeMRgFiXNweHycORsOdOGx8nsgBfx3ZkrW1PIDUkKZ+W2bI5llzN2YHSL+5VW2cnKq+K6qUkoisJtV6fyn6/s5M0NJ3nkhmEX7Za4v+QwpfYy7h16Z5OWMoD80joOny7lhqsGXNb/7cjo4azL/oy12RvJiBra5FojUiL4aPNptuTs5aOzqxkUnsaMxCmN24sr7fz27f3UOz0kRgYQHeZHSmIIQX4mgv3NBPqZiQq10Sc6sN0xCiGEEEKI7qfHFHKarvFZ7pf0DUokOaT5LnUrt51l9fYcUhPTGRI2nOPaJrJrtnPQU8Ug/XrGDYxmT2Yxx3MqGNK/aVc5l9fNytPr+OL8VmL8onhg+D0kBjZ0m7s1/XqsRgtf5G7F4XEy2n8mJ89XccusFGaOSmDF7sNsqHoX6gNJdEwGXeHQ6TLcHk1aQi4hyi8SBaXV4+SSYoPwtxo5fKbsooXc7syGCVTGDmxYqy482Mq1U/rz7udZ7D1Rwuj0qGaP03WdT7I/J8YvimERg5rd55Nd5zAaVCYPu3TheTGqojK//9X88+hb7Cs6yOiYjAu2JySCOW0P754uIz4gljsH3txY7JVU2vn92/twub387LaRUqwJIYQQQvQiPaaQO1ByhFJ7GUsGzG+2JSW7sJq1O84xaUgM31/Y8OF7vj6Qz3O3sOr0ev5n5zPckLwYW4CLLcfO0ifejIKCqqioikJJfRmvH3+PgroipiVMYsmA+ZgN33bbUxWV65MXYTPaWHt2A4ccRQQGDGPa8DjqPfUc8Kwl0GIlvGoGH3yew97MCkwGlSA/EykJIZ2Vpk6l1VeCakC1Xl4BYTaYCLeFUdCKJQgAVFVh8NfLEOi63mLL2s5jRfSPDbpgco+ZoxPYfrSQtzaeZFC/MPysTV8CR8qOk19XeEHR9F0FZXVsPVzAoslJBPm1bv22i8mIGkpcdgxrszcyMno4qqJS77azNnsDm89vxxBgIKJmFD+96gYMasOSAaWVdn739n4cLi+PL82QIk4IIYQQopfpEYWcrutszNlMpC2c4ZGDm2z3eDX+ueY4gf4mls5KaXxcVVRm9ZnGoLA0Xj/+Hq8efxsGwRHgZ1ubXifIHMiDw7/P4PDmx9ApisKC/ldTVwubSzYQORQ8jOWVw29Q6armkYz76R/Uh6+OFfHOxlPU2t1cNSIOVb3yulV6cg9h3/gioGMZfR2mwTNR1PavSxbrH0VRK5cggIZxcruOF5NbXNtsEVNQVse5olqWzki+4HGDqnLX3HT+5/U9LPvyDLfNTr1ge0Nr3BeEWUMZHT2i2Wt//OUZzCYDN81KxWV3tTrmlnzTKvfKkTfYXbgfr+5lxel11LnrmRQ3Fi0/jU0nSnDN1bFZoKzKwe/e2Y/d6eHxW0bQN0aKOCGEEEKI3qZHFHJZlWfIqcnl5tRrm20hWb09m/MldTx8/TD8rU0nv4gLiOGJUQ9xoOQIp4vK2LD7HNNGxJEQ5Y+OjqZrqIrKmOgMAsz+l4wn73gUimM4pQmH+OWO31DvsXPXoKUkBfcFYMLgGAb3D+PzveeZPPTyut51N7qu4z7yKc6v3kUNS0CxBePc8Tbuk9uwTrkTQ9SAdp03xi+aY2Un8Wrexlanixny9QQjh8+UNVvI7T5ejAKMaabrZf/YIGaOTOCzveeZMCSGpLhvJ6LJqjzD2eocbk5d0mwcZwuq2XuihMWT+xMcYKHEB4UcwPDIwcQHxPLG8ffR0UkK7suNqYvpE5jASf9KPttVzOEzZSTHB/O7d/ZR5/Dw+NIR9IuRSXSEEEIIIXqjHlHIbTy3mQCTP+NjRzfZdq6ohjU7cpgwOJoRKREtnsOgGhgVPZwRkRrbNm2jOjeUq0YNaXMsZwuqOXymjOunTSJ2wFBeO/Yu8/rNZGzMyAv2C/Izs2RKUpvP353pXg/ObW/gztyMsd8orNPvBaMFz9ndOLe/Tf3y/8E08CosY29AsVy6IP6uGP8ovLqXUkc50X6XHlMYHGChT3QAR86Us2BCvwvj1HV2Hi8iNTGE0EBLs8dfOzWJvSdLeG19Jv/9vdGNk998kvMFgeYAxseOafa4DzedJtDPxOwxiW26v0tRFZXrkheyLGsNM/tMZUx0RmOX0eT4YAJsJrYcKuDjL89Qa3fzHzePkJlQhRBCCCF6sW5fyOXXFnKkLJMF/a++YMwaNHSp/Mea4/jbTNwyK7WFM1zIoKqMTo9i66EC7E4PNkvbUrB6ezb+ViMzRiZgsxgZEj6wSVxXIt1Ri33DX/AWnMCcsQjz6GtRvm4dNSWNxZgwFOeeZbiPbsCTvRfLuJsxpkxs9SLoMf4NE48U1hW1qpCDhu6V63eew+70cLzqGPuKDhIfEIvFG0pBdTkzRw1r8Vibxcits1J4YdkRNuxuWFsupzqX4+Unm4yP/MbRs+Ucz6ngllkpbX7etEZ6WAo/G/uTJo+rqsLw5HC2HS7Eajbw2M0jGBAn68EJIYQQQvRm3b6Q++zcl5hVE1MTJjbZtnZHDrnFtTx03dBLrif2XeMGRvPFvjwOnCplwpCYVh+XW1zL/lOlLJ7cv/GDfG8o4rwVedjX/wm9vgLr9PswpTT9v1DMNqwTb8WUOgnH1tdwbPo76rHPMI9YgLFvRmPR15IYv28KuWKGt3KSzyH9w1izI4cjZ8tYXraKenc9B0qOoKNjy4BP6nZz7EA8iYHxDAxLJTX0wm6fI1MjGZEcwcdfniEhyp8dtV9gM9qYHD++ybU0XefDzacJD7Jy1Yj2LQB+OaYOjyMrr5rvzx8oi3oLIYQQQggu/um6i1W7athdtJ8JcWMIMF3YVS+3uJZV27MZNyiakaltm94/OSGYsCALO4+3brr7b6zano3VbGDW6IQ2HddT6Y5a3Ce3Ub/8f8DjxG/hU80Wcd9liOiL3+L/xDL1bnR7NY5P/0L9B7/Anfklutfd4nFWo5UQSzAFbZjwZEB8MDaLgS05+6l0VvG9wbfy+ym/wnpuChG1oxgSMZBqVw0bz23mz/tf5qNTq/Bq3sbjFUXhngUDiQ334y9rtnOg5AjTEiZiM1qbXGvviRJyCmtYMqU/JmPnv2xSEkL49X3jSU6QIk4IIYQQQnTzFrkdBXvQdI0ZiVMvePybWSr9rQ3d49pKVRTGDoxmw+5cau3uVrXm5ZfWsTezmPkT+jY7ocqVQKuvxFtw4us/J9EqzgOghvfBNucR1ICma+81R1FUzOnTMKVOxnNmN66Da3B8+U+UPR9jHjob08DpKGZbk+Ni/aMpqm99cW00qAzqG0amax2hASEMjRjI2fwaKgr9uW7UaCYNbJhoxuV1s/z0Gj7P3UJ29TnuGXxb40LyATYTjy0dwS83/A2n10Ci0nTcpMer8fHm08RH+jNhcOtbcIUQQgghhOgol1XIrVq1ihdffBGPx8Ndd93FbbfddsH248eP84tf/IK6ujpGjx7Nr371K4zG1l9yT+F+RkYNI8IWdsHj63aeI6eohgeXDCGwnet4jRsYzfqd59hzorhVXeVW78jGbDL4fJKLzqZ73ei15Wg1pWi1peg1pWg1pXhLzqBXfV1EmawYopMxJ4/DEJuOIao/itr2p4qiGjAlj8c4YBzevKO4DqzBufN9nPtWYew/EjU4FjU4uuFPUBQxflFsy9/ZOItoayT21TlWVcqI0JmoisrO40UYDQoZKd+20poNJm5KXcKA4H68lfkhv9n9Z7436BYGhjeMq3QrtXgCz2OtTOLlj7N47OaAC1q+th4uoKjCzsPXD7sil5IQQgghhBA9T7sLuaKiIp599lk+/vhjzGYzS5cuZdy4cSQnf7tu1xNPPMH//M//MGLECH7+85/z/vvvc+utt7b6Gk6vk1l9pqFpOnmldZzOqyIrr4qdx4oYkx7F6PSo9oZPn+gAYsL82HWs6JKFXFF5PTuPFTFnTJ92F45dSfe4cHz5L3IKT+CtrQD0bzcqCop/GGpYAsb0qzDEpqFG9L2sNeH+naIoGBOGYEwYgrf4DK5D6/HmHsZzctsF+4VGhOEKMVJy/DOiB13dqnOXm0+gayqm6n5oms7uzGKGJoU3u9D3qOgRJATE8fcjb/DCwX8wr99M5vWfxcZzX6IoCo9ctYSXSk7zx/cP8PjSDJLignC6vazYepbk+GCGJ7euRVIIIYQQQoiO1u5Cbvv27YwfP56QkBAA5syZw/r163nooYcAyMvLw+FwMGLECACuu+46nnvuuTYVciFKNB+sLeF0/mkcroaxTUF+JkamRnL77KazVGr2atyH1qM7a9G9XtC8oHlA86JrHlBUrJPvRA0IR1EUxg2KZuXWs1TUOFucph4aWuOMBpU5Y3tea5zudWPf8Dze3MMEDJmCyxKGGhiBEhDe8Ld/aLta29rLEJWEbdaDDbG57GjVxWjVRWhVRcRWZoOeS+6BZUQlT0Ix+130XPVuOwfLDmCpS+RkST0nYiupqnUxblDTteO+Ee0fxROjf8x7J5axNnsjWVXZnK3KZlzMKPqGRfHELUH89u19PPPeAZ64ZQTHsiuoqnXxwOIhrZ6BUwghhBBCiI7W7k/wxcXFREZ+230tKiqKQ4cOtbg9MjKSoqK2TS7SJ7eWg7VOxg+OITk+iOT4YCJDbM1+oNZqSrGv/QNadQmKLRBUAxiMDS1LqhFUA1ppNu7ML7GMvhaAcYOiWbH1LLuPFzF7bJ9mY9h2uIAdR4qYMTKe4ICWi73uSNc8OD57EW/uISxTvkfU1EWUlNR0dViNFLMNQ0RfDBENC6n3cdfBll9RZNRxn96FeeBVFz1+Z+FeXJqbkYEj2Xmyki0H87GYDAwf0PJ6ggAWg5k7Bt7EgJB+vH9yBV7Ny9V9pwEQFmTliVsy+O1b+3nm3QNoOgwbEE5qYogvblkIIYQQQgifaHchp2naBQWVrusX/Hyp7a0xz3uOxUkQde2jqM1MjvENV8k5Clb/H7gcxN3xK6yJA5vdL//NX+LN2UPE3DtQFIXIyEAGJASz91Qpty0YfMG+TreXlz8+xIZd5xg6IIJ7lgztUd0qdc1L8fI/4cneR/js7xM8Zj4AkZGBXRxZyyIJJNgSSGmQCme2Ezl1UYv7arrGtl1fkRLen3l9M9i2aztfHStiakY8CfEhrbrekqhZjOw7kNL6CgbHfbt4e2RkIL95aDI/++s2yqrs/GDJ0Bbz1p3z2VNJTn1L8ulbkk/fk5z6luTT9ySnviX59J12F3IxMTHs2bOn8eeSkhKioqIu2F5SUtL4c2lp6QXbW8M8+jrqP/8b5/71C2xzH0X1C2myj7coi/r1z6IYTNgW/owaawI1LbQ66YkjcW99naKTxzGENXSTHJkSwQdfnObIySKiQxu68hWV1/PCsiOcL6ll4cS+LJ7cH0edE0eds03xdxVd13BsegXPqe1Yxt2Mq98USkpqiIwM7FYtcs2JskVy1u2h5tRJCk+ewBAa1+x+x8pOUFBbzPf63EJkgBmLyYDT7WVEUnib7tFGEImmoCbHGICf3TaSwvJ6Akxqs+fsCfnsaSSnviX59C3Jp+9JTn1L8ul7klPfkny2jaoqhIcHtLy9vSeeOHEiO3bsoLy8HLvdzqeffsrUqd8uExAfH4/FYmHv3r0ArFix4oLtrWFKmYhtziNolYXUL3sab/n5C7Z7zh2ifvXvUCwB+F3zCwzhFx/DZuw/GhQFz+ldjY+NTW8YT7XrWEO3z92Zxfzq1d1U1Dj4yY3DuW7qAAxqt15u7wK6ruPc8hqeU9sxj74W8/B5XR1Sm8zsM5VSzcHbMSE4Mr9scb/N57cTaA4gI2ooJqPKoH6h+FuNDO4f1uIxbRUaaGFg31CfnU8IIYQQQghfaXeFEh0dzaOPPsqdd97JkiVLWLhwIcOGDePee+/l8OHDAPzhD3/g17/+NXPnzqW+vp4777yzzdcx9hmB3zU/A12jfsX/4jl/FAD3qe3YP/kzamgsfot/gRp06UXBVVsQhriBuM/sRtcbZm4MD7aSkhDMV8eKeGvDSV5cfoT4CH/+391jGTag82cp1HUdz7lDaPVV7TrWuf0t3JmbMY9YiDnjmg6IsGMNjRjETalLyPQ382HpLrRmFhEvtZdxtCyTyXHjMH49Ucvts9P46a0ju2SxbiGEEEIIITrbZU1XuGjRIhYtunAc09///vfGf6enp/Phhx9eziUAMET0w2/Jf2Ff/yz2dX/EmDwOz6ntGOIGYpv9cLOLS7fEmDQW55ZX0cpzMYQ3THAyblA0b356koKyeq4enciN0wdgNHR+QaBrGs4db+M+uhE1LBG/xb9AMVlbfbxr1we4j27ENHQO5jHX99hZFqcmTKC8+AQbOEbEoXdZkHHHBdu/zNuBoihMjh/f+FhooOWiM48KIYQQQghxJekxzRdqQHhD98n4gXhObcfYfzS2uY+2qYgDMPYbCYp6QffKcYOiGZEcwYNLhnDLrJSuKeI8LhwbX8B9dCPGfqPQKs7j2PRKY8vhpbgOrsV1cC2mQTOwjF/aY4u4b1wz/DZG1ntZW3GY7fm7Gx93eV3syN/N8MghhFiCL3IGIYQQQgghrlydt4CYDyhmG7a5P8FbeApDTBpKO8aufbd75TetVv5WEw/fMKwDIm4dzVGDff2f0IrPYJlwK+ahs3EdXIdz53u49q/EMnLxRY93n9yKc+f7GJPGYpl0e48v4gBUg4ml4aOpKdvJO5kfEmQOYEjEQPYUHaDeY2da/MSuDlEIIYQQQogu02Na5L6hqEaMcQPbVcR9wzhgLHp1EVrZOR9G1j5adTH1K/4Hrewc1lkPYh46GwDTsLkYkyfg2rMMd/a+Fo/3nDuAY/M/McQPwjr9XhSlx/2XtsiaNpXbC6qINQTwjyNvklOdy+bz24nzjyE5pH9XhyeEEEIIIUSXuXI+9beBqd+ohu6VZ3ZdeucO5C0+Q/3y/w/dUYttwZOYksY0blMUBevUu1Ej++P44m9NZuyEhqUX7Bv+ihqeiO3qH6MYTJ0ZfoczhMVji+jPPaUOAs0BPLf/b5yvzWdawsQrotVRCCGEEEKI9uqVhZxiDcAQPwj36V2tHoPWEm9JNrrH1ebjPDn7qV/1GzBZ8V/8nxhjUprGaTQ3TOZitGD/5M/ojtpvr1uR37B+nn8otnmPtXmsYE9hSpuMf1keP+wzB4NiwGa0MiZmZFeHJYQQQgghRJfqlYUcgClpLHpNCVppTrvP4T69i/pl/w/H5y+h61qrj/PkH8f+6V9Qw+LxW/yfqCGxLe6r+odim/1j9LoK7J+9iK550WrLsK/9A4pqwG/+Y6i2oHbfQ3dnGjAODCbCso/wxOgf80jG/VgM5q4OSwghhBBCiC7Vawu5htkrDe3uXqnVluHY8i8UWxCe7H249q1q3XHVxTg2vIAaHIPfgidQ/S4986IhOhnrlLvw5h3FufUN7OueQXfVY5v3GGpQVLvi7ykUiz/G/qNwZ31FhDmQxMD4rg5JCCGEEEKILtdrCznFGoAhYTDuM23vXqlrGo7PXwZdx2/xf2JMmYhr7zI82fsvfpzLjv2T59B1Dduch1HMfq2+piltCqYhV+PO3IRWVYxtziMYIvq2Ke6eypQ2FVz1eC4y6YsQQgghhBC9Sa8t5ABMSWPQa0rRSs626TjXgdV4C09inXQHalAU1infQ43sj/2Ll/FW5Dd7jK5rODb9Ha0yD9usB1GDY9ocr2X8UkzD5mKb/TDGuIFtPr6nMsSlowSE4z6xpatDEUIIIYQQolvo1YWcsd9IUA2429C90luUhWvvcowDxmNMaVjLTDGaG2aNNJqxf/pndGddk+Nce1fgyd6HZfxSjAlD2hWvohqwjl+KsU/XrXnXFRRFxZQ6GW/eMbTasq4ORwghhBBCiC7Xqws5xeKPIWEInjO7W9W9UnfZsX/+MkpAGNYpd14wBb4aEIb16ofQa0qxf/4yuvbt5CfuM7tw7VuBMXUypiGzO+RernSm1MmAjvvktq4ORQghhBBCiC7Xqws5+Hr2ytoytOLTl9zXse0N9NpSrNPvb3Z8mzEmFcvE2/HmHsK15yMAvKU5ODa9gvr1hCWy/ln7qEGRGOIG4j6+Cd3r7upwhBBCCCGE6FK9vpAz9ssA1Yj7zO6L7ufO+grPqe2YRy5uds23b5gHTcc08CpcB9bgOroR+6fPoVj8sV390BW3YHdnM49YgF5Xjvv45q4ORQghhBBCiC7V6ws5xeyHMXHo190rm18LTqspwbHlNQzRKZgzFl3ynJaJt2OITsG57U10ezW22Q+j+oX4OPLexxA/GENsGq79K9Hdzq4ORwghhBBCiC5j7OoAugNj0hg8OfupX/7/oZj9UIxmMFka/jZa8BacAMA64z4U1XDJ8ykGI9arH8Lx+UuYBs3AENm/o2+hV1AUBfOYG7Cv/F9cRzdiGbGgq0MSQgghhBCiS0ghBxj7jcKYPAHdXoXucaLbq9E9LvA40T0NLT/WafegBka2+pyqXzB+C3/aUSH3WsaYFAyJw3AdXIt50PRWrcXnLTyFVlWIKW1KJ0QohBBCCCFEx5NCDlBMFmwz7u/qMEQrWcZcT/3Hv8R1aD2W0ddddF+tspD6dX8Etx01vE+vWURdCCGEEEJc2Xr9GDnR8xgi+mJMGoPr8Kdo9uoW99PdDuwbnmvoDmv2w7V3eecFKYQQQgghRAeSQk70SObR14LHievAmma367qOY/M/0CoLsM56EPOwuXhy9uMtPtPJkX4dj9uJtzIfz/kjePIzuyQGIYQQQghx5ZCulaJHMoTEYUyZhPvYZ5iHzkENCLtgu/vQejxndmMZdxPG+EEYIvvjPvwpzr3L8Jv3WIfGplWX4DryKXpNKVptOXptGbqz9oJ9zCMXYx61RNYVFEIIIYQQ7SKFnOixLKMW48nagWv/SqxTvtf4uCfvGM5d72PsPxrTsHkAKGYbpuHzce16H2/hKQwXWQvwcuiueurXPYNeW4oaHIMSEI4hKgklIBw1IAwlIBzPyW249q0AzYt5zPVSzAkhhBBCiDaTQk70WGpgJKb0abiPb8Y8fD5qUBRabRmOz15EDYnFOu37FxRJ5sEzcR9e39Aqt+BJn8ej6xr2z19Gry7BtvBJjLFpze5niEkBVcV1YDW65sEy7mYp5kSn0r0e8LpaNeurT67nqEWrr8AQltgp1xNCCCF6AynkRI9mHnkN7hNbce5djnXK97BveB7d68Hv6h+jmG0X7KuYLJiHL8D51Tt48jMxxqX7NBbX3uV4zx3EMun2Fos4AEVRsUy+CxQD7kPrQfNimXCrFHOiU+geF/a1f8BbeKphJtf4gRjj0jHEpDV5zVz2tbxu3Ec34ty3Ctx2rDN+iGnAOJ9eQwghhOitpJATPZrqF4Jp8Ezch9Zjd9SilZzFOvth1JDYZvc3DZqO69A6XHs+xrDoZz4rntxn9+LatxJj6hRMg2Zecn9FUbBMuh1UA+4jn4KmYZl0G4oi8w+JjqPrGo5Nr+AtPIlp0Ay0inzcRzY2fKGgqKiR/THGDcQQNQAlIAw1IBws/m1+nei6jufsbpw7P0CvKcGQOAxcdhyf/w3FYMbYL6OD7lAIIYToPaSQEz2eZcQC3Me/wJt7CHPGIkz9Rra4r2I0Y85YiHPbm3jzjmFMGHzZ1/dW5OHY9HfUyCSsk+9o9YdeRVGwTLiloZg7tK6hZW7KnVLMiQ7j2v0xnjO7sIy7CfPw+UBDC523KAtv/nE8+cdxHVwHuvfbg4xmVP+wb8d5BkaghsShhsajBkehqBf+GvEWZeH46l20oizUsASs8x/HmDAE3WWnfs3vsG98Advcn2BMGNKZty6EEEJccaSQEz2eYg3AOukOvCVnMI+69pL7m9Kn4TqwFueejzHED7qsVjndWYf90+dQjGZsVz+EYjS36XhFUbCMuwlFNeA6sBp0DcvUu6WbpfA51/FNuA6sxjTwqsZJgKDhyw1j/CCM8YOw0LD+olaRj1Zbhl5bjlbXMPOqVluOJ/cwen3ltydVDKgh0Y2FnVZZgOfMLhRbMJapd2NKnYKiNnwxoZht+M17jPrVv8X+yXPY5j920S7IQgghhLg4KeTEFcGUOglT6qRW7asYTJhHXoNzy6t4cw9h7DO8XdfUdQ37F39Dry7FtvDJJksgtJaiKJjHXA+A68BqjP1HtTsmIZrjOX8E59bXMSQOxTLp4q3GismKISoJQ1RSs9t1txOtsgCtIg+tMh+tIh9vWS6e7L2gmhqW1hg+D8VkbXpuawC2BU9gX/Vr7OufxW/BExiiBvjsPoUQQojeRAo50SuZ0ibjOrAG555lGBKHtasFrGLze19PbnLHZbcsKIqCefQS3Fk7cO5f1e6YhPh33vJc7BueRw2NxzbzQRTVcFnnU0wWDJH9MET2u+Bx3eMCXUcxWS56vGoLwrbgSepX/h/1a5/Bb+FPMUT0vayYhBBCiN5IBuOIXklRjVhGXoNWmo0nZ3+bjtV1Hdexz6nc9iGmtCmYBs3wWUzm4fPQirLwFmT65Jyid/PUlGNf9yyKyYpt7qM+n5XyuxSj+ZJF3DdU/1D8Fj6JYrI2zKBZkYeuaehuZ8NSBXUVaNXFeCvy0OqrOixmIYQQoieTFjnRaxlTJqIcWIPzq3dR/cOatDA0R3fW4djyKp4zu7EljcBwiW5qbWVKm4pr30pc+1djjBvos/OK7k2rKUV31DQsIn+JYkuzV+MtOIG38CTewpPoLgdqYARqYCRKUCRqYCRqYASKXzCFK55Hd9bhd83P2931t6OogZH4LXyS+pW/pv6DX1xkRyO2+Y/7fLkQIYS4FK2qCD20aTdxIboLKeREr6WoBqxT7sLx2YvUL/sVpoFXYRlzPYo1oNn9PYWncHz+EnpdJeaxNxEz60ZKS+t8G5PRjHnYXJw738dbfKbFcUriyqDVluHauwL3ya2gawAotiDUkFjU4GjU4BjU4Fh0Vz3ewhN4C06iVRU2HGwwY4gegBoUhVZTiufsHnRn7YUXUFRscx7ptl0X1eAY/K75Ge5T20ExgMGEYjQ1/G0wgcGIa+9y7Bv+gv+S/0YNju7qkIUQvYTn3CHs65/lfHgsxol3yeRMoltSdF3XuzqIlpSV1aJp3Ta8HicyMpCSkpquDqPb0Z11OPcux330MzDbsIy5AVP6tMbZ9nRNw7V/Fa59y1ECI7HN+CGGqKQOy6fuslP7zuMYY1KxzXnE5+fvznrLc1Srr8S1fzXu45uAhvUNDbHpDd/+VhWgVRWhVRWi26u/PcjshyEmBUNMGsbYVNSIfiiGC7+L0112tJpStJoS9JoSwpLSqPXv13k31gG06mLqlz2NYg3Ab8l/oVj8uyyW3vL87EySU9+SfPqGVlNC3cf/D9UWjKp78FQVY0q/Csu4G7v0PehKIM/RtlFVhfDw5hsYQFrkhECx+GOdeBum9Kk4t72Jc+truDM3YZ10B4p/KI4v/oa34ATG5AlYJ9/ZoeOMoGGadvOQq3HtXY63PBdDWGKHXk90Ht1R27Ag/ZEN4PVgSpuMeeTihoW3m9vfWYdWVQRGE2po/CXXGFTMNgzhiRjCG54ztshAanv4L0w1KArr7B9jX/M77Bv/im3eo03WrhNCCF/RPS7sG14AXcM25xEi+ySQ98kbuA9/gufcASyTbsfUf3RXhykEIIWcEI0MYYnYFj6F5/ROnF+9S/2K/wGjBRQF61X3tnp5A18wD56F69B6XPtXY5v5QKddV3QMXddxH/4E574V4HJgTB6HZdQS1OCYix6nWPyley1gjE3DOuV7ODb/A+e2t7BMvlNmdRVCdAjn9rfRSrOxzn64oYu72Yp1/FJMA8bj+PJfODY8j6ffSCyT7kD1D+3qcEUvJ4WcEN+hKAqm5PEY+47AtX8V3rJcrBNv6/SxOYo1APOgGbgOrUMbdS1qyMU/8IvuS3fZcWx6BU/2XgyJw7CMu1FaWdvBlDYFrbIA18G1qKFxmIdc3dUhCSGuMO4TW3BnbsI8YgGmfiMv2GaI7Ifftf/d8KXcnuV43v85ptSJGBOHY4hLRzGauyhq0ZtJISdEMxSTFcvYG7s0BtPQObiObMB5YA22q77fadfV3U5ch9bjydqBMXUS5mHzmozF8vk1XXZcB1ZTlzIUPSTtimlt0SoLsX/6HFpVIZbxt2AaOvuKubeuYB57A1pVIc4db6MGRWPsM6yrQxJCXCG8ZedwbH0dQ9xAzKOva3YfRTVgHj4fY//ROHe+jztzS8P4eoMZQ/xAjInDMPYZhhoYCYDudaNVFaNV5qNVFjSMfa4uwZg8DtOgmfL7QFw2KeSE6KZUv2BM6VNxH9uENmoxamBEh15P1zU8p3bg3P0hel0Falgirt0f4cnaiXXq9zBEJ3fIdbX6Suzrn0UrzaHowBrU8L6YRy7C2G/kJceEdWeenAPYP38ZRTU0TJ8fP6irQ+rxFEXFOv0+6lf+H/bP/orf4v/EEJbQ1WEJIXo43VmHfcPzDWPmZz6Aohouur8aFIXt6ofQPS68BSfwnDuI59xBnOcO4tzWMCOvrmvoNSXwnTkFFf8wFLMV57Y30SoLsEy4rXFiNSHaQwo5Ibox8/D5uI9twnVwHdbJd3TYdTz5mTi/egetNAc1sj/WmQ9gjEnFk7Mfx9Y3qF/xv5gGTccy9gYUs5/PrqtVFlC/7hl0ezXW2Q8TaPZSuuVDHBueRw1NwJyxEGPS2A75Rac5alCMFp93h9H1r2c53bMcNbwPttk/7vAivDdRTFZsc35C/bJfYV//LH4LnrjkWEMhhGiJrmvYv/g7ek0ZfoueQrUFtfpYxWjGmDgUY+JQ9Im3oVcV4jl3CE/eUVSTFTV5fMNyMiGxDeuEmqzomoZz53u4D3+CVlOGbeYDKCZLB96huJLJ8gO9iEz56ludlU/H5n/gztqB/y1/QPUL8em5taoinDvfx5O9F8U/DMvYGzAmj7+gJUx32XHu+Rj3kY0ofsFYJt+Bqd+oy762tygL+/o/gaJgm/to45IOxUWVeM7swrVvFVplPmpwDOaMRRj7ZfikiNSqS3DuW4Hn1DYwWTElT8CUNqVhOv8Wurnouo5Wfh5P1g7cZ3aD5kUNikQNikYJikINikINjkKxBePc9gae7H0Ns5xOvbtLx01cya95b0k29nXPoOsatlk/6pQWzys5n11Fcupbks+20XX96y/ePsYy8bZmx952VE5dRzfi3P4WanhfbHN/4vPf792VPEfb5lLLD0gh14vIi8e3OiufWlUhde//DNPQuVjG3QTOOrTaMvS68oa/a8vRHbUotiAU/1BU/zCUgNCGLhzWABRFbVhfrLKgoZ9+RUNffW9lPnp1MRjMmEcswDxsDoqx5W8FvcVncGz5F1pZLsZ+IzGPXNzuhabd2XtxfPYSin8YfvMfQw2KAi7Mqa5reM7uaSjoynMBUALCUcMSMIQlooYloIYlooZEt2o6eq22HNf+lbgzt4CqYEq/Ct1Zh+fsHvC6UcMSMKVNwZg8ofEbWa26BPfpr/BkfYVWkQeKiiFhCIo1AK26GL26+MK13gAUFcv4pZiGXN3l4x+u9Ne8Vl2M/ZM/oVUWYZl0O+ZB0zv0eld6PruC5NS3JJ8Xp2setNJzeAtP4i08hbfwJLqjBmPS2IYulc28Z3dkTj05B7B/9iKKNaDhC81e0FVcnqNtI4WcaCQvHt/qzHzaP3sRz5k9YDCAx3XhRtWIYg1At9eA7v23bQYUsx+6o+aCx9TgaNSQuIbiZeBVrf4mUNc8uA9/inPvcvC4UCP7Yxp4FaYB41BM1ladw3Xsc5zb3kCN6N/wLeR3urE0l1Nd1/DmZ+ItOYNWdh6t/DxaZcG396oaUUPjMUT0RY3ogyGiH2pYYmNXFa2+EteBNbiPfwG6jil9GuaMRY3TRuuuetxZO3Gf+BKt5CyoBox9hqPZq9GKsgAwxKRiTB6Psf/oJt1uGhbhLvm6sCtpWLS7g8YTtlVveM3rLjv2z17Em3sI05CrsYxfesnxLe3VG/LZ2SSnviX5bKBrXvS6CrSaUvTaUrSqIrxFWXiLTzf+DlUCIzHEpmKIScWUPKHF3hMdnVNvaTb29X9CdzuxXf0QxoTBHXat7kCeo20jhZxoJC8e3+rMfGrVxTj3LEOxBaEGhKH4h6EGhKMEhKPYAhta3XQN3V7d8Murrhy9rqLhj6MWJTASNTQWQ0gcSlDkZS+orDvrcJ/ajvv4poaWKpMVU8pETOnTLmil03Ud3VmLXl+JXl+F59wh3Ec+xdBnOLaZDzYZF9DanOpeD1pVAVpZLlr5ebxl59BKc74tWBUFNTgWNSQGT+4R0DyYUidjHrmocTax5njLc3Gf2Ion6ysUWyDG5AmYBozrsWPcestr/rtjTgyJQxvGnDTTDVfXNfTqEnRnLYaoAW2+Tm/JZ2eSnPpWb8mn7nE19EqpKUOvLUP7+o9eU4pWW4peWw669u0BioIa1qfhi7bYVAzRKa1eA64zcqrVlmFf9yxaZQHmjAWYBs9q9Vi9b7r+K2brRX+/dRe95TnqK1LIiUby4vEtyefXv0CKsnAd34TnzK6GLorhfUA1NhRv9irQLmwlNKVfhWXyHc22mlxOTnVdb/jFXnoOb2k23tIctIo8DNHJWEYt7pUTYvS256jr+CacW99ADY7GNvthdM399fMhB60sB2/pOXDbAbDOfADTgHFtOn9vy2dnkJz61pWcT29lPs4tr6NV5jft0o6C4h+CGhCBEhiBGvj13wFf/zsgDMVgatd1Oyunuqsex+Z/NnT3NxgxpUzCNGwOhpC4ZvfXHDV4Tm3HfWILWvl5ANToZEwDxmMcMLZNk7Z0piv5OdoRpJATjeTF41uSzwvpjlrcWTvwnNkNBhOKXwiqXzCKX8jXf4JRA8Iu+o2h5NS3emM+PfnHsW94Hpx13z5oMKOGJ37d/bZvQ0tydTH+N/5vmyYY6I357GiSU9+6UvPprcjDvvq3oOsNE18FhDcUbQHhX/dOCb3sniYt6eyceivzcR/6FPepbeB1Y+gzHPOwuRhi00HX8eYdxX3iSzzZ+0HzNAxxSJuC7qpvGM9dfv7r8dyDG4q6fiNRzLZOi/9SrtTnaEeRQk40khePb0k+fU9y6lu9NZ9adTHuUzsaZhaN6Ncw7fd3lrDwVuZT/9EvMcQPwjbnJ62elKa35rMjSU5960rMp7f8fEMRpxqwLXyyxRaqjtJVOdXs1biPfYH76EZ0Rw1qeCK6sx69tgws/l8PZ5iKISzxguO85bl4snbiztrRsK/BhCltKpax1/t0+aD2uhKfox3pUoWcrCMnhBDiiqIGRWEZtbjF7YaQOCxjb8S5423cJ77EnD6tE6MTQrSWt+wc9jW/B9WA38KnUEN6Txd51RaEZdRizMPn4c7agfv4JtSQWEzjbm5olWyhq6ghLBHD2ETMY65HK8rCfXIL7uOf48nei2XS7Zj6j+7kOxEdSQo5IYQQvY5pyCw8Oftx7ngHY/ygHjFJgBC9ibc0h/o1v0MxWvBb+FPU4OiuDqlLKEYz5vRpbf7CSVGUhsldYlIwpV+FY8u/cGx4Hk+/kVgm3o4aENZBEYvOpF56FyGEEOLKoigq1mnfB8Cx6RX0785w10m85bnUr/0Dji9fxZO9H93t7PQYhOiOvCXZ1K/+LYrJit+ip3ptEecrhqgk/K79JZZxN+HJPULdBz/HdXQjutb573vCt6RFTgghRK+kBkZgnXArji//ifvIBsxD57TrPLrmxZt/HENsOoqhdb9WvaU52Nf8Hh0db1EW7sxNYDBiiBuIMXE4xj7DUYOklVBcWXRdx5Ozv2GJF2tA42QlDTNLhqP4BaOVZFO/9vcoFv+GljhpLfcJRTViHj4fY//ROLa8hnPbm7hP7cB61fc7fdyh8B0p5IQQQvRaxrQpGLL34tz1IYbEoW3+QKPrGo5Nr+DJ2oEanYxt5oOX7LLk/eaDqtGC/6KnUPzD8BaexJNzAE/uQZzb38S5/U3U0DhMg2ZgSpva4mLFQvQEuq7hyd6Pa98KtLJzKLYgdM174eyyAF/PPKn4h35dxPXMNTy7MzUoCtv8x/Fk7cC5/W3sa5/B/4b/r1tMhCLaTgo5IYQQvZaiKFin3k3dB7/A8cUr+C3+RbNrHDZH13WcW9/Ak7UDY/IEPDn7qf/4l1hn/BBjwuBmj/EWn6Z+7R8aWhsW/LSx1c0YPwhj/CDgVrSqQjznDuI+vQvntjdx7V+NecQCTOnTpKATPYqua3jO7sW1byVaeS5KcDTWq+7FmDweRTWgu+wNC3nXljYu7q27nZhHzEcNCO/q8K9YiqJgSpmIGhxN/Yr/xbHtTWzT7+vqsEQ7SCEnhBCiV1P9QrBOvgvHZ3/FdWANlpHXXPIYXddx7nwP9/EvMI9YgGXsjXgr83FseB772j9gHn0t5oyFKMq3Q9E9haewr3sGxRbU0NrQwgdVNTgG89AYTENm480/jmvfCpzb32oo6IbPxzToKhSjxWf3L0R7eMvP4zq4DsVgALMfisW/4Y/ZD8Xih+6oxXVgLVrFedTgGKzT78M4YNwFX5QoZhuGsAQIS+jCO+m9DFEDMGcswrVvBe6+IzAlje3qkEQbSSEnhBCi1zMNGIsney+uvStA82IeMf+ixZJr30rch9ZjGjQT85gbgIZlDfyW/DeOLa/i2vMx3qIsbNPvQ7EG4Ck4gX3dH7/tMuYfesmYFEVpbKnz5Gc2FHRfvYPr4BrMw+ZiGjQTxSQFneh83qIs6tf9EXQNxWRFd9aB191kPzUkFuuM+zEmjbtgLUfRfZhHLsKTexjHltcwRKe06r1JdB9SyAkhhBCAdfKdOHS94dvpE1uwjLupoQXh3xYMdx36BNfeZRhTJ2GZdNsF2xWTFev0+3HHpOLc/jZ1H/8S8/B5OL96HzUwAtvCJ1H9QtocmzEuHWNcOp7Ck7j2rsC5831cB9dhHj5PCjrRqTznj2D/9DkUv1D8FjzeOBmJ7nGhu+rRnfXgrEPXNQzRKVLAdXOKasQ2/T7qPvpvHJv/gW3eY03e80T3Ja8uIYQQAlAs/thmPYht0c9QrIE4Pn+J+pX/i7f4TOM+ruObcH71Dsb+o7FOveeCrpON51EUzINm4HfNzwFwbnsTNTgK26Kn2lXEfZcxJhW/BU/gt/g/USP64tz5PnXvPoHr0Dp0jyxfIDqW+8wu7OufRQ2Kxu+an10wo6RiNKP6hWAIjcMQk4IxNk2KuB5CDYnBMmEp3vNHcB/7rKvDEW2g6Lqud3UQLSkrq0XTum14PU5kZCAlJTVdHcYVQ/Lpe5JT35J8tp+uabhPbsG1+yN0ezXG1EkE9x9I2Sf/wJA4FNvsh1u11IDuqMV9YgvGtMmo1kCfx+ktPIVz73K8eUdRbEFfj6Gb3mPG0Mlz1Lc6Mp+u45twbn0NNWoAfnMfRbH4d8h1upve8hzVdR37+mfx5h/H7/pfddiSBL0ln76iqgrh4QEtb+/EWIQQQogeQVFVzOnT8L/5t5iHz8eTtZOyT17BEJuK7eqHWr1enGINwDx8XocUcQCGmBT8FjyB7ZpfoIYl4vzqXereeQLP+SMdcj3ROzkPrMW55VUMCUPwm/9EryniehNFUbBOuwfFaMHx+d/QNU9XhyRaQQo5IYQQogWK2YZl3E343/R/hE2/Hducn3TLJQCMjQXdz1FsQdg/+TOevGNdHZbo4XRdaxiPuet9jEljsc1+RMZjXsFUvxAsU7+HVpqNa9/Krg5HtIIUckIIIcQlqEFRhEy8FsVs6+pQLsoYk4pt4U9Rg6Kwf/InPAUnujok0cPomoYnPxPHtjepe/sxXAfXYhp4FdYZP2x1S7TouUz9R2NMnYRr/yq8hae6OhxxCfKKFEIIIa4gqjUQ24Insa/6Nfb1z+I3/3EM0cldHZboxnTNgzfvOJ6ze/Hk7EO3V4PBhDFhCMZxNzc7e6u4clkn3k5dwQnq1/wec8ZCzMPnoRhMXR2WaIYUckIIIcQVRvULxrbwp9Sv+jX1657Bb8FPMUT26+qwOpTurMO5fxWKyYp5xAL54HkRurMOb8lZvMVn8Bafbmh5cdWDyYoxcRjG/qMx9hmGYrJ2daiiCyhmG36LfoZzxzu49nyM++Q2rBNvxdhneFeHJv6NFHJCCCHEFUj9evHx+pX/R/3a3+O38KcYwvt0dVg+p2sa7hNfNsww6qgFdDxn9mCd/gMMEf26OrxOo7sd2LOz8ZTXgK6BpqHrGujehn87a/EWn0UrPo1WVdh4nBoSi6n/KIx9R2JIGNwtx4CKzqcGhGO7+iE854/g3PYm9vXPYuybgWXCrahBkZc+gegUsvxALyJTvvqW5NP3JKe+Jfn0rZ6aT626mPpVvwGvG9uipzCExnd1SI0uN6feoiwc295EK83GEJOKZeJt6PUVOL58Fd1ejTljEeaMRVf02C7NXo37yAZcRz9raFW7CMUWhBqZhCEqCUPUAAxR/VHMfp0Uac/UU1/3vqR7PbgOf9IwAYquYR6xAPPw+e0q+jsin7quX7Fdfy+1/MCV+84mhBBCCNSgKPwWPkn9yl9jX/1bbPMewxDRt9XHe8tzce54p2HMVNwgDAmDUEMTuvSDk1ZfiXPXB3hObkPxC8E6436MA8Z/HVNf/G/8Xxzb3sS1bwWenP1Yr/pBp7dG6poXAEU1dMj5tepiXIfW4z6xBbwejP1GEjFuLtUOBRS14Y+qNv5bMVlR/EOv2A+8ouMoBiOWEQswJU/A+dW7uPYux5t3DNvCp7p00XetvhLH5n+iVeZjm/cfHbb2XXcmLXK9iHyr5FuST9+TnPqW5NO3eno+vRX52Ff/Ft1R2zCBwSVaqnRdw31kA85dH6CYbGDxQ68qAhpadgxxgzDED8QYPxg1MKJdMbUnp67MzQ2FpdeNeeichvtoYTZRd/ZenFteQ3fWYR65GPOI+Siq77/D1h21eMtz0cpy8ZblopWfQ6vIA6MFy5jrMaVf5bMPvN7SHFwH1+I5swsUFVPKpIa1CkNie/xztDuSnDblzvwSx5f/xDLhFsxD57TpWF/l05NzAMfmf6C7nY1LYtgWPIkhPPGyz92dSIucEEIIITCExuF34//g3P5WQ0tV9j6sV32/2XFkWl0Fjk2v4M07iqHPcKzTvo9qC0KrLcObdwxP3jG8eUfxnP4KJ6AERWNMGIwhYQjGuIEdtkyD59wBnF++iiEuHevku1BDYi66v6nfKAwxqTi3voFrz8d4cvZjm/FD1ODodl1f17xoVYVoZblo5ecbize9rrxxH8UWhBqWiGnwLLSSszi3vo77+CYsk27HGJParutCQyHu2vUBnpz9YLJiGjoX89DZqP6h7T6nEO1hTJuCIXsvzt0fYeybgRoUdcljdF3HueU1srP3gC0YxT8U1T+soZX463+robGogRcff6d7XDi/eg/3sc9QwxOxzXgARVGpX/M76lf/Br95j2GISvLVrXZ70iLXi8i3Sr4l+fQ9yalvST5960rKpyd7P46trzWMIxs+H/OoxY2zPLqz9+Lc/C90jwvLhKWYBk5vtjuerutoFfkNBV3eUbz5meBxgqI2jL9KGIwxYQhqVBKK0nxrVFtyqlUVUrfsV6iBUfgt/kWbx+e4T+/CseVV0DWsk27HmDLpkt0MdV1rmJL/3MGG4q0yD7yeho2KoeGDZ1gihvBE1PA+qGGJqH7B3zlex3NmN86v3kWvK8eYPAHLuJvaVHxp9ZW49izHfWIzGC2Yh8/DPHgWisW/yb5X0nO0u5CcNk+rLaPug19giErCNv+JS76WnPtX49r9IX6pY3C5vGh1Feh1Fej1VcC3n/XVsESM/UZi7DcSNbzPBef1luXi+PwltIo8TEPnYBl7Q+P7llZTQv3q36E7arDNfRRjbFqLseiaF8/pnWhVRZgzFnbrGW4v1SInhVwvIm9GviX59D3JqW9JPn3rSsun7qzDseNtPCe3oYbGYZl0B56sHbgzv0SN6It1xv1tGnOiez14i7Lwnj+CJ+8oWkk2oGPsPxrrrAebLeZam1Pd7aB++dPo9dX4XffLS35r3xKttgzHF3/DW3AC44BxWKfc1exkH7qu48nZh2vPcrTyXBRrIGpEX9SwBAzfFGwhsa2eREV3O3EdWI3r0DpQDJhHXoN56OyLfoDUXXZch9bhOrQeNC+mQTMwZyxCtQW1eMyV9hztDiSnLXMd+wLn1tewTL0bc/q0FvfznDuIff2fMA4YR+LNj1NaWtu4Tdc86PVV6LXleIvP4Mne+/VC5DpKQPjXRd0otLJzOHe9j2L2wzr9PowJQ5pcR6urwL7md2g1ZdjmPNxkH93japjh9tB69JpSAAyxadhmP9zsFyPdgRRyopG8GfmW5NP3JKe+Jfn0rSs1n55zh3BsefXr7oFKQ4vP6Osue6ZH3VGL68inuPatxDz6Oiwjr2myT2tyqus6jo0v4Mnei23e4xgTBl9eXJrWUFTtXY4SEIZt+v0YYlIar+XNPYxzz8dopdkowdFYRi5umEjFB2PctOpinDveaegeafZDDYxEDQhD8Q9DCQhr/LdWnotr7wp0Rw3GpLFYxt7Qqu5rV+pztCtJTlum6xr21b/DW5qD/03/12xLs1ZZSN3yX6EGRuK3+BdExUZcMp+avRpPzn482fvw5h1tbAX/bjfvix1rX/t7tIoCrLMexNRvJLqzDtexz3Ef/hTdUYManYxlxAJ0lx3H5n+gBkdjm/sf7R7r25GkkBON5M3ItySfvic59S3Jp29dyfnUXfW4Dn2CIX7QRbsktfm8uo7ji7/hydqBbc4jGPtmXLC9NTl1HliLa9f7WMbdhHn4fJ/F5i3Kwv75y+i1ZZhHLsYQnYxz7zK0oiyUwIiGAi5lYofMOuk5fwTPmT1odeXoteVodeVNlg4wxKZhGXdzm8b7XMnP0a4iOb04raqIug//C0P8IGxzHrmgK6Tusje0pDtq8bv2l6iBEW3Op+6y4zl/GFAw9h/dqllXdWcd9eueQSvJxpg8AU/2XnA7MCQOwzxiAYaY1MbzePKPY//0ORSjBdvcR9s0o29nkEJONJI3I9+SfPqe5NS3JJ++JflsH93jon7l/6FVFeK35L8xhH7bXfNSOfWcP4J93TMY+4/BOvMBn0+dr7vsOLa+jidrBwCKfxjmjEWY0qZ0+tpzusv+9bihcjCYLviw2VryHPU9yemluQ6tw/nVe1hn/BBT8njg69a6T57Dm3sI24InMMYNBDovn7rLjv2TP+EtPIkxaSzm4fNbLNK85XnY1z2D7qrHNutHGBOHdnh8rSWzVgohhBCiyyhGM7bZP6b+4/+H/dM/47/kv1s1HkWrLsH+2YuoIfFYp32/Q9Y/U8w2bDPux90vA91Riyl1crsWOfZVLAazDUJ731pYomczDZmN+/QunNvfwhA/CNUW1LDW3LkDWCbe3ljEdSbFbMO24El0R+0FExA1xxAWj9+S/8K+/o/Y1z+LderdmNKmdFKkl6frVvETQgghRK+gBoRjvfoh9OrShu6MmnbR/XWPE/uGv4CuY5v948Z1ojqKKWks5kEzuqyIE6InU1QD1mnfR3fV49z+Nu6ze3DtW4kpbQqmwTO7NK5LFXHfUP1D8Vv0cwxxA3Fs/gfOvcvp6k6LuqahVRdddJ92t8jl5+fzxBNPUFZWRv/+/fnDH/6Av/+F37Dl5eWxcOFC+vTpA0BERAT/+Mc/2ntJIYQQQvRQxtg0LJNuw7n1dVx7PsYy9oYm++heN97zR3Ad2YBWlott7k/aveabEKLzGMISMGcswrV3OZ7sPahRSVgm39khLekdRTHbsM17FMeXr+LauxzdWYdlwq2dcg+624lWnou37Bxa2bmv/z6PMSAYHnqpxePaXcj96le/4tZbb2XBggW88MIL/PWvf+WJJ564YJ8jR46waNEinn766fZeRgghhBBXCNPA6WilObgOrEaN6AORMxuLN/eZ3Xiy94PbDhZ/LBNvw9hneFeHLIRoJfOIhXjO7kF31GK7+sfden22liiqEeu07+M0++E+8il43Fim3NniWpjt1bBcyym8ecfwnD+KVnoWvmkBNPthiOiLadB0zHGpFz1Puwo5t9vN7t27eeGFFwC47rrruP3225sUcocPH+bkyZMsXryY4OBgfvGLX5CW5rvZsIQQQgjRcyiKgmXS7Xgr8nBseoWigkPUndzbWLwZ+4/GNGAMhrhBnT7ZiBDi8igGI37X/AJ0rduuy9YaiqJgmXALitGM68BqdK+rYZzuZcxgq2saWmU+3vNH8eQdxVuQCR4XKCpqVBLmEQsxRCahRvRpWI7k61ZAVb14a2C73iUrKioICAjAaGw4PDIykqKipn04LRYL11xzDUuXLmXLli386Ec/Yu3atZjN0gddCCGE6I0Ugwnb1Q9Rv+xp7KcPYEoajTFpDIb4QSiqFG9C9GSK2dbVIfiEoigN3b+NZlx7PsbhcWGd8cNLfsGkux1olYVolflolQVf/ylEqyoErWE9PDU4BlPqFAwJgzHGpaOY/dof56WWH1i3bh2//vWvL3isb9++nDt3js2bNwPg8XjIyMjg8OHDF73YNddcw+9+9zvS09PbHbAQQgghej7N7URRDdLyJoTo1qp2raZsw7/wSx5F1PWPo/7bpEieqhJqM3dQd/wrnHknvt2gqJhCYzCFx2OKiMcckYit72CMwZE+i+2S757z5s1j3rx5FzzmdrsZN24cXq8Xg8FASUkJUVFRTY594403WLhwIaGhDSu967re2IrXGrKOnG/JWii+Jfn0Pcmpb0k+fUvy6XuSU9+SfPqe5NS3emQ++0/DMlmjfutr5L75P9hmP4xur8Zzdg/uM7vRSs4AoIb3wTxyMWp4ImpIHGpQVOMXVRrgABwuoA333yHryJlMJkaPHs3atWtZtGgRy5cvZ+rUqU322717Nw6Hg3vvvZddu3ahaRpJSUntuaQQQgghhBBCdDrzoOkoRjOOza9Q987j6I6GYkyN6It57A2Y+o9GDY7p9Lja3Z/hl7/8JU899RQvvvgisbGx/PGPfwTgnXfeobi4mEceeYRf/OIXPPXUU6xYsQKLxcIzzzyDqsrSdUIIIYQQQoiew5Q6CYxm3Mc+x5AwFFPSaNSgpj0SO9Mlx8h1Jela6Vs9sjm7G5N8+p7k1Lckn74l+fQ9yalvST59T3LqW5LPtrlU10ppHhNCCCGEEEKIHkYKOSGEEEIIIYToYaSQE0IIIYQQQogeRgo5IYQQQgghhOhhpJATQgghhBBCiB5GCjkhhBBCCCGE6GGkkBNCCCGEEEKIHkYKOSGEEEIIIYToYaSQE0IIIYQQQogeRgo5IYQQQgghhOhhpJATQgghhBBCiB5GCjkhhBBCCCGE6GGkkBNCCCGEEEKIHkYKOSGEEEIIIYToYaSQE0IIIYQQQogeRgo5IYQQQgghhOhhpJATQgghhBBCiB5GCjkhhBBCCCGE6GGkkBNCCCGEEEKIHsbY1QFcjKoqXR3CFUdy6luST9+TnPqW5NO3JJ++Jzn1Lcmn70lOfUvy2XqXypWi67reSbEIIYQQQgghhPAB6VophBBCCCGEED2MFHJCCCGEEEII0cNIISeEEEIIIYQQPYwUckIIIYQQQgjRw0ghJ4QQQgghhBA9jBRyQgghhBBCCNHDSCEnhBBCCCGEED2MFHJCCCGEEEII0cNIISeEEEIIIYQQPUynF3LPP/88CxYsYMGCBfzud78DYPv27SxatIjZs2fz7LPPNjnmySef5OOPP278OT8/n9tuu425c+fywAMPUFdX12nxdze+yCfABx98wFNPPdUpMXdnvsjn3r17ueGGG1i8eDF33XUXeXl5nRZ/d+SLnO7Zs4frrruORYsW8cMf/pCqqqpOi7+78dVrHuDYsWMMGTKkw2Pu7nyR02XLljF58mQWL17M4sWLmz2mt/BFPouLi7nvvvtYsmQJS5cu5fz5850Wf3dzufksKytrfF4uXryYGTNmkJGR0an30N344jl6/vx5brvtNhYvXswdd9zRq3/X+yKfhw4d4vrrr2fRokXcf//9lJSUdFr8PZreibZt26bffPPNutPp1F0ul37nnXfqq1at0qdNm6afO3dOd7vd+j333KNv2rRJ13VdLyws1O+//3592LBh+kcffdR4nvvuu09fvXq1ruu6/vzzz+u/+93vOvM2ug1f5NPhcOi///3v9REjRug//elPu/J2upyvnp/Tp0/Xjx8/ruu6rn/wwQf6D3/4wy65n+7AVzmdNWuWfurUKV3Xdf33v/+9/swzz3TJ/XQ1X+VT13W9vr5eX7p0qZ6amtoVt9Jt+CqnTz/9tL5q1aquuo1uw1f5vOuuu/S3335b13Vdf/vtt/VHHnmkK26ny/nyNa/ruu71evXbb79dX7lyZWffSrfhq5w+/vjj+ltvvaXruq6//vrr+mOPPdYl99PVfJFPTdP0adOm6Tt27NB1XdfXrFmj33///V12Tz1Jp7bIRUZG8tRTT2E2mzGZTAwYMIDs7Gz69u1LYmIiRqORRYsWsX79egBWrVrFzJkzmTdvXuM53G43u3fvZs6cOQBcd911jfv3Nr7I5+7du9E0jSeeeKKrbqPb8EU+XS4XjzzyCOnp6QCkpaVRUFDQJffTHfgipwBr164lOTkZt9tNUVERQUFBXXE7Xc5X+QT4zW9+w1133dXZt9Dt+Cqnhw8fZtmyZSxatIjHH3+817Ya+yKf5eXlZGZmsnTpUgCuv/56fvKTn3TF7XQ5X77mAT766CNsNhuLFi3qzNvoVnyVU03TqK2tBcBut2O1Wjv9XroDX+SzoqICh8PB+PHjAZg+fTpbt27F5XJ1yT31JJ1ayKWkpDBixAgAsrOzWbduHYqiEBkZ2bhPVFQURUVFAPzgBz/gxhtvvOAcFRUVBAQEYDQagYYn0Df79za+yOfkyZN58skne+0b0Hf5Ip9ms5nFixcDDW/yzz//PLNmzeqcG+iGfJFTAJPJxIkTJ5g2bRo7d+5kwYIFnRJ/d+OrfH722Wc4HA7mzp3bKXF3Z77KaWRkJA8++CArV64kNjaWp59+ulPi7258kc/c3Fzi4uL4zW9+w/XXX8/DDz+MyWTqtHvoTnz1/ATwer289NJLPPbYYx0ed3fmq5w+8sgjvPrqq0yZMoV//vOf3HvvvZ0Sf3fji3yGhobi5+fH1q1bAVizZg1ut5uKiorOuYkerEsmOzl16hT33HMPTz75JImJiSiK0rhN1/ULfv53zW2/2P69weXkUzTli3y6XC4ef/xxPB4P999/f0eG2yP4IqdpaWls376dBx98kEcffbQjw+32LiefJSUlvPjii/zXf/1XZ4TaY1zuc/SFF15g1KhRKIrCD37wA7Zs2dLRIXdrl5NPj8fDUmtuJwAACB1JREFUsWPHGD9+PB999BEzZ87s9WO4ffEeumXLFvr160daWlpHhtpjXG5Of/rTn/L000+zZcsWfvWrX/HQQw+h63pHh91tXU4+FUXhueee4+WXX2bJkiXU1NQQEhLSa7/AaYtOL+T27t3L9773PR577DGuvfZaYmJiLhjQWFJSQlRUVIvHh4WFUVNTg9frbdX+V7rLzae4kC/yWVdXxw9+8AM8Hg8vvvhir38jutycOp1ONm7c2PjzNddcw4kTJzo05u7scvO5adMmKisrGwfpAyxevLixi1BvdLk5ramp4dVXX238Wdd1DAZDR4bcrV1uPiMjI/H392f69OkALFy4kEOHDnV43N2Vr37Pb9y4kfnz53dkqD3G5ea0vLycM2fONPa4mTNnDiUlJb22BckXz1Gj0cgbb7zB8uXLueaaa9A0jZCQkA6OvOfr1EKuoKCAH/3oR/zhD39o7Bo1fPhwzp49S05ODl6vl9WrVzN16tQWz2H6/9u7t5Co9jaO418dt5Za2VRokbnNdC6UThjdZNFB0rB0AsFSKewmsC6LMUqNEG1SE5Ui0owswQ7kGEwqVnS46CBmIJVdFJKGUkReeCLNfRF7Xtrt13ht8p3J3+du/vPwX888rGF45v9fa/3xB9HR0djtdgDq6urGjf+dOaOe8h/OqueBAwcICQmhpKQEb2/vyUjdZTmjpl5eXhw9epT29nYAbt68ycqVKyclf1fjjHomJyfT3NyMzWbDZrMBYLPZ8Pf3n5TP4GqcUVNfX18qKip49uwZABcvXiQ2NnZS8nc1zqjnokWLCAoK4u7duwDcuXOHyMjIScnf1Tjzd76trY3o6OhfnbLLc0ZNZ8+ejY+PDy0tLcDXRsbPzw+j0Tgpn8GVOOscPXTokOMPm6qqKuLi4vD01FPSfsRrMg9WWVnJ8PAwBQUFjrGUlBQKCgrYv38/w8PDrFu37ofXbeTk5GCxWDh9+jTz58+nuLj4V6fukpxVT/nKGfV8/vw5t27dYsmSJZjNZuDr3vCzZ8/+8vxdkTNqajAYOHnyJNnZ2YyOjhIYGEheXt5kpO9y9J13PmedoyUlJeTm5jI0NMSff/7puAX3VOOsc7SsrIycnBxOnDiBv7//N/NNJc78zr99+5agoKBfma5bcEZNPTw8KC8v59ixYwwNDeHn50dZWdlkpO9ynHWO5ubmkpOTw+DgICaTacr+zv+vPMam8oZeERERERERN6Q1SxERERERETejRk5ERERERMTNqJETERERERFxM2rkRERERERE3IwaORERERERETejRk5ERFxaV1cXJpOJtLS0796zWCyYTCY+fvw4obnLy8sdD5y3WCxUVlZ+F9PQ0MCqVavo6en5ZtxqtZKWlsbo6OiEjv1vXr58yZo1a5w2n4iI/L7UyImIiMvz8fHhzZs3dHd3O8YGBgZobW39qXkfPXrEyMjIuDFxcXHExsaSlZXF30/saWlpwWazUVRUhMFg+KkcAEZGRjh//jx79uyhv7//p+cTEZHf36Q+EFxERGQiDAYD8fHx3Lhxg7179wLQ1NTExo0bOXfuHAC1tbVUV1fj6enJ3LlzOXLkCKGhoVgsFvz9/eno6KCnpweTycTx48epq6ujvb0dq9XqaMaePn1KSkoKHz58IDw8nKKiInx9fTl8+DBms5mamhqSkpKwWCzk5+cTGBhIa2srhYWFDA4O4unpyb59+1i/fj0DAwPk5ubS2dnJp0+f8PPzo7CwkMWLF5Oens6sWbN4/fo1O3bsYNmyZXR0dFBeXk5GRsb/rc4iIuI+tCInIiJuISkpCZvN5nhdV1eH2WwG4OHDh1RUVHDhwgXq6+tJSEggMzPTsYLW3t5OZWUldrud7u5uGhoaSE1NJSoqioMHDxIbGwtAb28vVVVVNDY20tvbS1NTEwC+vr4UFxdTWlpKVlYW8fHxrF27lr6+PrKysrBarVy/fp1Tp06Rm5vLu3fvuHfvHjNnzqS2tpbGxkaioqK4dOmSI/+ZM2dit9tJT09n6dKl5OfnM2/evMkqp4iIuDmtyImIiFuIiorCYDDQ3t7OnDlz6O/vJyIiAoD79++zZcsWjEYjANu3bycvL4+uri4AYmJi8Pb2BiAiIoK+vr5/PcamTZuYPn06AOHh4d9cexcZGcnOnTtpbm6muLgYgLa2Nt6/f09mZqYjzsPDg46ODuLi4ggODqa6uprOzk4eP37MihUrHHHR0dHOKo2IiExBauRERMRtbNu2jfr6eoxGI4mJiY5xDw+P72LHxsYc179Nmzbtm9i/V+r+ycvLa9y44OBgFixY4IgbHR0lLCyMK1euOGJ6e3sxGo3U1NRw+fJlUlNT2bp1KwEBAY7GEr6u8omIiEyUtlaKiIjbSExMpKGhAbvdTkJCgmM8JiYGu93uWEG7du0aAQEBhISEjDufwWD44c1OxrN8+XI6Ozt58uQJAC9evGDz5s309vby4MEDzGYzycnJhIaGcvv2bafe4VJERKY2rciJiIjbCAwMJCwsjBkzZhAQEOAYX716Nbt372bXrl18+fIFo9HImTNn8PQc///KDRs2UFxczOfPnyeUj9FopLS0FKvVyvDwMGNjY1itVhYuXEhGRgbZ2dlcvXoV+Nr0vXr1akLHERER+SePsf+2v0RERERERERckrZWioiIiIiIuBk1ciIiIiIiIm5GjZyIiIiIiIibUSMnIiIiIiLiZtTIiYiIiIiIuBk1ciIiIiIiIm5GjZyIiIiIiIibUSMnIiIiIiLiZv4C+TGNqWY6nEAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x648 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Merge DAX cumulative log returns with the cumulative log returns of the equally weighted portfolios and plot\n",
    "dax = dax.rename(columns={'cumulative_log_return': 'cumulative_log_return_dax'})\n",
    "long_short_dax_ew = pd.merge(long_short_ew,dax[[\"cumulative_log_return_dax\"]],left_index=True, right_index=True)    \n",
    "long_short_dax_ew[[\"cumulative_log_return_long_ew\", \"cumulative_log_return_short_ew\",\"cumulative_log_return_dax\"]].plot(grid=True,figsize=(15, 9));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2wAAAIZCAYAAADa9LyXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzdeXycZbn/8c+smUwmk33f07TpvhfaAi17BdlEQBABBREOKkfO0aO/IyiIuCCK6FFERVFRQAHZ96VlKXRP96bZ92Umy2yZyWRmnt8fadKGZpkkM5lJcr1fL1/SzDPPc8/TJJ1r7vv+XipFURSEEEIIIYQQQkQddaQHIIQQQgghhBBieFKwCSGEEEIIIUSUkoJNCCGEEEIIIaKUFGxCCCGEEEIIEaWkYBNCCCGEEEKIKCUFmxBCCCGEEEJEKSnYhBBCCCGEECJKaSM9AICuLheBgLSDC4WUFBMdHc5ID2NGkXsaWnI/Q0/uaWjJ/QwtuZ+hJ/c0tOR+hp7c0/FRq1UkJcWN+HhUFGyBgCIFWwjJvQw9uaehJfcz9OSehpbcz9CS+xl6ck9DS+5n6Mk9DR1ZEimEEEIIIYQQUUoKNiGEEEIIIYSIUlKwCSGEEEIIIUSUioo9bMPx+310dVnw+byRHsq00t6uJhAIRHoYM0q03FOtVk9SUhoaTdT+2AohhBBCiBCL2nd+XV0WDAYjcXGZqFSqSA9n2tBq1fh8kS8uZpJouKeKouBy2enqspCamhXRsQghhBBCiKkTtUsifT4vcXFmKdaEAFQqFXFxZplxFkIIIYSYZaK2YAOkWBPiBPLzIIQQQggx+0R1wTbb7N69k6997SujHvPhh+/z5JOPA/Dcc0/z3HNPT+n1w+XQoQP89re/Cus1Ivn6hBBCCCGEmIio3cMmhnfkyKHB/77ssisiOJLQqq2toaurM9LDEEIIIYQQIqpIwRYERVF4+OFf8957m9FqNVxyyeW899673HjjV1i5cjUtLc18/eu38PTTL3LffXdjMMRy9OgRnE4HX/nKV3n99VeorDzKGWecyde/fgevvPIie/bs4rvfvRuAr33tK9x449CZnz17dvH73/+W3l4PDoeT22+/g9zcfJ5//lkAMjOzaG1tAcBsTqCxsZ477vgfAH796wdJT0/n4os/wy9+8VOqq6sIBAJce+31nHfep4J6zfX1ddx//304HHYMhli+8Y1vsmDBItrb2/jBD+7C4bBTXFxCWdlu/v3vV0Y8zyuvvMirr76EzdbNaadt4Morr+ZnP/sRbW1tqNVqbrnlq8yfv5A//vF3uN1u/vKXR0lLSx/x/jz88K/w+wMUF88hKysbq9VCQ0M9bW2tXHTRpdxww02Ten333Xc3cXEmyssPY7Va+OIXv8yll16G0+nkhz/8Ho2NjWRn52CxtPGjHz1AVlb2sOf/3//9Fuef/ynOPPMcAG688Qucd96nOHz4ID/4wY+pr6/j85//LC+88DrJySn81399jZtv/g8WLFgU1PiFEEIIIcTsMC0Ktg/3t/DBvpawnPv0pVmctmT01L13332b/fv38te/PonP5+O2276M19s74vFWq4VHHvkzr776Ej/+8T088cSzxMTEcNllF/KlL90c1LieeeYpvvOduygoKGTXrh089NAD/PWvT3HppZcD8OlPX8Kjjz4CwLnnbuLGG6/l9tv/G41GxZYt7/DII3/mL395lNLSBdx55z24XE5uvfVGFi5cTE5O7pjXv/feu/jCF77Ixo1nc+DAfu6889s88cSzPPTQA5x99nlcfvmVbNnyLm+++dqY57JY2nn88X+h1Wr5/vf/H5/+9CWcfvpGrFYrt912E4899g++/OVb2bNnFzfccBOvvPLiiOdqaKjn6adfwmQy8eijj1BZWcFvf/tHnE4HV111GZdffhXx8fETfn0A7e1t/Pa3f6S6uoqvf/0WLr30Mv785z+Qn1/AT37yC44cOcQtt3xp1PNv2nQhb775KmeeeQ4NDfV4vV4uueQynnrq7yiKwu7dO0hKSmbPnt2cfvoZNDTUM3/+wjHHLYQQQgghZpdpUbBFWlnZLs4++zz0ej16vZ7HHvvHqHuh1q5dD0BGRiZFRXNISkoGwGw243DYg7rmXXfdy9at7/Puu29x8OB+3G73iMcmJSVRUjKX3bt3YjDoyc8vICUllZ07t9Pb6+Hll18AwOPxUFNTPWbB1tPTQ2NjIxs3ng3A4sVLMJvN1NfXsWPH9sGZr40bz8JkGrs4mjdvPlpt/7fazp3bqaur449/7C82fT4fTU2NY55jQF5eASaTafDPK1euRqfTkZSUjNlsxuVyjlmwjfb6AE455VRUKhXFxXOw223Hxr2N733vhwDMn7+Q4uI5o15j/frTefDB++npcfHWW6+zadMFxMWZyM8voLKygl27dnLVVddQVrYbozGWlStXS6iIEEIIIYQ4ybQo2E5bMvYsWDhptVpOfC/d0tI85M21z+cbcrxOpxv8b41Gc9L5VCoViqIM/tnv9510zFe/ejMrV65ixYpVrFq1hnvuuXPUMW7adCHvvPMmer2O88+/AIBAwM9dd91Lael8ADo7OzCbE0Y9D4CinNxzTFHA7/ejVqsJBJRhnjWymJiYwf/2+wP86lcPD47DarWSlJRERcXRwWNGuz8nngtAr9eP+LyRjPb6+s8ZM3i+Af2vO/hebDqdjtNOO4MPPniPd955k5/97CEA1q07nR07tlFfX8s3v/kdbr/9VtRqFevXnxH0uYUQQgghxOwhKZFBWLZsJZs3v4PP58Pj8fDf//11TKZ4amqqAHj//c3jOl9CQiJ1dTUoikJzcxOVlZVDHrfbbTQ01HHTTbeydu1pvP/+lsFiQaPRDBYWJzrjjI2Ule1m27aP2bDhLABWrlwzmCJptVq54YZraGtrHXN8cXEmsrNz2LLlHQAOHNhPZ2cHxcVzWL36lMFlkB999CFOp2Ncr33VqtU8++y/AKipqeb66z9Hb69nyOsa6/5M1mivbySrV586+Lqrqiqprq4ac0Zs06YLefLJx0lISCQzs/8Dh/XrT+f555+hsLCIhIRENBotH374PmvWnBqiVyeEEEIIIWaSaTHDFmkbN57FkSOHuPHGawkEFK688hrmz1/Afffdzcsvv8AZZ5w5rvOtXn0KL7/8PNdc81kKCgpYunT5kMfN5gQuuuhSrrvuKrRaLStXrsHj8eB2u1m+fCX33Xc3ycnJQ54TE2NgyZJl+Hw+jEYjADfeeDM///lPue66qwgEAtx22+1B7V8D+N737uVnP/sRjz76CDqdnvvuux+dTsc3vvFN7r33+7zwwrOUlMwLaknkie6443+4//77uOGGq1EUhbvu+gFGYxwLFiziT3/6PQ8//GtuuumWUe9PKIz0+kbyxS/exI9+dA833HA12dm5pKSknjTb90lLly7H6XQOSfMsKChEURRWrFgFwIoVq6ipqRr8OxNCCCGEEOJEKiWYNWRh1tHhPGmZXWtrHZmZBREa0fSl1arx+YJfujde//rXk6xefQpFRcWUlx/hpz/9IX/60+Nhu1400GrVvPzyS2RlZbN06XJaW1v5+te/wlNPPYdaPbWT1DPh5yItLR6LZXwzs2J0ck9DS+5naMn9DD25p6El9zP05J6Oj1qtIiXFNOLjMsM2y+zdu4cHH/zZsI898MBDpKamjfr83Nw87r77u6jVKvT6GL797Tt5++03+NvfHhv2+Mce+8dkhzwuk319IykoKORnP/sxgYAflUrNt771v+zfvzcs1xJCCCGEEGKAzLDNMOGeYZuNoumezoSfC/nULfTknoaW3M/QkvsZenJPQ0vuZ+jJPR2fsWbYJHRECCGEEEJEhQ53J0c6KyI9DCGiihRsQgghhBAiKjxd8SK/2fsobT2WSA9FiKghBZsQQgghhIg4j6+Xw53lBJQAL1S9FunhCBE1pGATQgghhBARd6iznL6Aj/lJcymz7KfGVhfpIQkRFaRgE0IIIYQQEbfXcgCTLo4vL/kC8ToTz1W9QhRk4wkRcVKwRZHdu3fyta99ZdRjPvzwfZ58sr/v2XPPPc1zzz09pdcPl0OHDvDb3/4qIte+4oqLaWlpnvDzH330Efbu3RPCEQkhhBCzS1/AxwHrYZamLiJWG8uFRedS2V3DwY4jkR6aEBEnBds0c+TIIVwuFwCXXXYFl112RYRHFBq1tTV0dXVGehgTsmfPLvx+f6SHIYQQQkxb5Z0VePy9LE9fAsBp2aeSFpvC81WvElAm11rHF/Bh6ekIxTCFiAhpnB0ERVF4+OFf8957m9FqNVxyyeW899673HjjV1i5cjUtLc18/eu38PTTL3LffXdjMMRy9OgRnE4HX/nKV3n99VeorDzKGWecyde/fgevvPIie/bs4rvfvRuAr33tK9x449CZrT17dvH73/+W3l4PDoeT22+/g9zcfJ5//lkAMjOzaG1tAcBsTqCxsZ477vgfAH796wdJT0/n4os/wy9+8VOqq6sIBAJce+31nHfep4J6zfX1ddx//304HHYMhli+8Y1vsmDBItrb2/jBD+7C4bBTXFxCWdlu/v3vV0Y8zyuvvMirr76EzdbNaadt4Morr+ZnP/sRbW1tqNVqbrnlq8yfv5A//vF3uN1u/vKXR0lLSx/x/jz88K/w+wMUF88hKysbq9VCQ0M9bW2tXHTRpdxww00jjmVg7G63G7VaxX/+57dYvLj/H4Y///kPVFSU4/F4uPPOH7Bo0eIR78F9992NzWajqamBL3zhi5SXH+anP/0hP/rRA8yZU3LSdW22bq677nM8++zLaLVaqqsrueeeu8jIyOQzn7mCdetO45FHfsPRo+X8/Oe/wmq1cscdt/G3v/0zqL8rIYQQYrorsxzAoDFQmjQHAI1awyVzLuDRA4+zrXU367JWT/jcz1a+xNbmHdx/xt3oNbpQDVmIKTMtCra+ox/SV/5eWM6tK92Abt5pox7z7rtvs3//Xv761yfx+XzcdtuX8Xp7RzzearXwyCN/5tVXX+LHP76HJ554lpiYGC677EK+9KWbgxrXM888xXe+cxcFBYXs2rWDhx56gL/+9SkuvfRyAD796Ut49NFHADj33E3ceOO13H77f6PRqNiy5R0eeeTP/OUvj1JauoA777wHl8vJrbfeyMKFi8nJyR3z+vfeexdf+MIX2bjxbA4c2M+dd36bJ554loceeoCzzz6Pyy+/ki1b3uXNN8dOcbJY2nn88X+h1Wr5/vf/H5/+9CWcfvpGrFYrt912E4899g++/OVb2bNnFzfccBOvvPLiiOdqaKjn6adfwmQy8eijj1BZWcFvf/tHnE4HV111GZdffhXx8fHDPvell55n/frT+fznr+fjj7eyb1/ZYMFWWFjM//7v93nmmad44om/8cMf/nTwHpxzzrmUle0dvAcACQkJ3H//gwC8/PIL3HjjV4Yt1vqPTWThwkVs2/YRp512Bm+++TqbNl2A0Whk164drFt3Gnv37qG9vQ2/38+2bVtZt+70Me+rEEIIMRP4A372WQ+yJHUBWvXxt6Yr0pZQYM7jperXWZW+bELFVqeniw+atuFX/FjcVnJMWaEcuhBTQpZEBqGsbBdnn30eer0eo9HIY4/9g+TklBGPX7t2PQAZGZkUFc0hKSkZozEOs9mMw2EP6pp33XUv1dWVPPbYH3nyycdxu90jHpuUlERJyVx2795JWdlu8vMLSElJZefO7Tz//DN88Yuf56tf/Qoej4eamuoxr93T00NjYyMbN54NwOLFSzCbzdTX17Fjx3Y+9akLAdi48SxMpuGLoxPNmzcfrbb/F/DOndv54x8f4Ytf/Dzf/Obt+Hw+mpoag7klAOTlFWAyHe8Ev3LlanQ6HUlJyZjNZlwu54jPXb36FJ544nHuvvu72O02PvvZqwYf27DhTACKiuZgs3WPeg8AFi5cHPSYAc4//0LefvsNAN599y3OO+9TrFt3Ort2baenp3+Ja0nJXI4ePcLHH2/ltNPOGNf5hRBCiOmqylaDq6+H5WlD/21VqVR8Zs6FdPfa2NL44YTO/VrtO/iV/m0L0ttNTFfTYoZNN++0MWfBwkmr1aJSHf9zS0szqhO+4PP5hhyv0x3/BEij0Zx0PpVKNST1yO/3nXTMV796MytXrmLFilWsWrWGe+65c9Qxbtp0Ie+88yZ6vY7zz78AgEDAz1133Utp6XwAOjs7MJsTRj0PgDLMWnFFAb/fj1qtJhAYX2JTTEzM4H/7/QF+9auHB8dhtVpJSkqiouLo4DGj3Z8TzwWg1+tHfN4nLV26nMcf/ydbt37A22+/wSuvvMgvf/lbYOjfk6Ioo96D4cYxltNP38D//d+DlJXtJiMjk7S0dAACAYXNm99hyZJlJCcns3PnDsrLD7N48dJxnV8IIYSYrsosB9CpdSxIKT3psblJc1iUMp/X695lffYpxOmMQZ+3w93JRy07WJu5mo9bd9LmkoJNTE8ywxaEZctWsnnzO/h8PjweD//931/HZIqnpqYKgPff3zyu8yUkJFJXV4OiKDQ3N1FZWTnkcbvdRkNDHTfddCtr157G++9vIRDoLyA0Gs2wARdnnLGRsrLdbNv2MRs2nAXAypVrBlMkrVYrN9xwDW1trWOOLy7ORHZ2Dlu2vAPAgQP76ezsoLh4DqtXnzK4DPKjjz7E6XSM67WvWrWaZ5/9FwA1NdVcf/3n6O31DHldY92fifrtbx/i9ddf5YILLuKOO77N0aPlIx472j34JI1GO2boiF6v59RT1/GrX/18sKCG/tnYv/zlUVasWMXKlWt45pmnWLRoybCFvhBCCDHTBJQAey0HWZhSSoxGP+wxl865AI/Pw+t174zr3K/Vvo1apebiOZtIikmkrac9FEMWYspNixm2SNu48SyOHDnEjTdeSyCgcOWV1zB//gLuu+9uXn75Bc4448xxnW/16lN4+eXnueaaz1JQUMDSpcuHPG42J3DRRZdy3XVXodVqWblyDR6PB7fbzfLlK7nvvrtJTk4e8pyYGANLlizD5/NhNPZ/+nTjjTfz85//lOuuu4pAIMBtt90e1P41gO99715+9rMf8eijj6DT6bnvvvvR6XR84xvf5N57v88LLzxLScm8oJZEnuiOO/6H+++/jxtuuBpFUbjrrh9gNMaxYMEi/vSn3/Pww7/mpptuGfX+TNRnP/s57rnnTl555UXUajV33nnPqMcP3IM//en3aLW6wXvwSaeeuo4HHvgxd955D0uWLBvxfJs2Xcjrr7/KmWeePfi19etP58knH2fp0uXExsbi8/Wxfr3sXxNCCDE71Nkb6e61cWnaBSMek2PK4pTMlWxp3MqZuaeRbEga87yWng4+bt3Fhpx1JMYkkGFMkyWRYtpSKVHQkbCjw3nSMrvW1joyMwsiNKLpS6tV4/NNLv52NP/615OsXn0KRUXFlJcf4ac//SF/+tPjYbteNAj3PR2PmfBzkZYWj8UyvplZMTq5p6El9zO05H6G3ky6p/+ufJl3Gz7gJ6d/D6MudsTjOj1d3PPxz1iZvpQbFl495nn/duif7Gov45513yEhxsw/jz7HtpZdPLDhB0O2tcDMup/RQu7p+KjVKlJSTCM+LjNss8zevXt48MGfDfvYAw88RGpq2qjPz83N4+67v4tarUKvj+Hb376Tt99+g7/97bFhj3/ssX9MdsjjMtnXNxm/+c1D7Nix7aSvz5+/gO98566wXVcIIYSYjhRFocxygNKkklGLNYBkQxJn553BG3XvkhWXwfkFZ414bHuPhW2tuzgr73QSYswApBvT8Ph7sXsdg18TYroIqmBzOp1cffXV/O53vyM3d+iSusOHD/Pd734Xl8vF6tWrueeeewYTAUX0WbZsxaSKqHXrTmPduqEBMKWl8znnnPMnO7SQmOzrm4yvfvU/I3JdIYQQYjpqdrVidXdwfv6ZQR1/UdH5dHq6eL7qVVSoOK9g+Oe9UvM2WrV2yOMZxv4PbNt6LFKwiWlnzNCRvXv3cs0111BbWzvs49/61rf43ve+x+uvv46iKPzzn9LsVwghhBBCjK6sfT8qVCxNWxTU8Rq1husXfI5V6ct4ruoV3qrfctIxra52drbtYWPuesz64/vsTyzYhJhuxizY/vnPf/L973+f9PT0kx5ramrC4/GwfPlyAC6//HJee23sRspCCCGEEGJ2K7McYE5iIfH6kffufJJGreGGhVezKn0Z/658+aSi7dXat9BpdJybv3HI1xNjEtCpdZIUKaalMdcu3nfffSM+1t7eTlra8T1BaWlptLW1hWZkQgghhBBiRmrvsdDsauWKuZeM+7kDRVsAhX9XvowKFefkb6DZ2cqutr2cV3DmSUWgWqUm3ZgqM2xiWprUZrNAIDAkaUdRlJOSd4IxXCpKe7sarVbaxE2E3LfQi5Z7qlarSUsbXyuFaDQTXkO0kXsaWnI/Q0vuZ+hN93u69fBHAJxdeiqpcRN7Ld9K+wq/+uhPPFv5EiaTgaMd1cRo9XxuxYXEx5z83rIgKZuqzrph7910v5/RSO5p6EyqYMvMzMRiOf5JhdVqHXbp5FiGi/UPBAJRE6U+VXbv3smf/vR7/u//fj/iMR9++D4NDXVcffUXBptiX3bZFYOPTyaCPpjrh8uhQwfYvPkdbrvt9im75n333c2KFau48MKLRz0ummL9A4HAtI/Jlajf0JN7GlpyP0NL7mfozYR7+kHtTvLjc1F6dFh6Jv5aPl9yJZ7ePv5a1v+e6FOF5+CxK3g4+ZwJmiTaXbtpbutCpz7+Fngm3M9oI/d0fMaK9Z/UtEFOTg4xMTHs2rULgOeff54NGzZM5pRiDEeOHMLlcgH9hdqJxdp0VltbQ1dXZ6SHIYQQQogw6/J0U2dvYEXakkmfS6PWcOOiz7MibQnxehPn5J0x4rEZxjQUFCw91klfV4ipNKEZtptvvpnbb7+dJUuW8MADD3DnnXfidDpZtGgR119/fajHGHGKovDww7/mvfc2o9VquOSSy3nvvXe58cavsHLlalpamvn612/h6adf5L777sZgiOXo0SM4nQ6+8pWv8vrrr1BZeZQzzjiTr3/9Dl555UX27NnFd797NwBf+9pXuPHGrwy55p49u/j9739Lb68Hh8PJ7bffQW5uPs8//ywAmZlZtLa2AGA2J9DYWM8dd/wPAL/+9YOkp6dz8cWf4Re/+CnV1VUEAgGuvfZ6zjvvU0G95vr6Ou6//z4cDjsGQyzf+MY3WbBgEe3tbfzgB3fhcNgpLi6hrGw3//73KyOe55VXXuTVV1/CZuvmtNM2cOWVV/Ozn/2ItrY21Go1t9zyVebPX8gf//g73G43f/nLo6SlpY94fx5++Ff4/QGKi+eQlZWN1WqhoaGetrZWLrroUm644aZR/x7/7/8e5MMPPyA1NZVAIMCKFasAeOSR37Br1w7sdjupqan84Ac/xmKx8K1v/Sd///s/URT40peu5Sc/+Tlz55YGdQ+FEEIIcbK9loMALEtfHJLzadQavrzkOvoCviEzZ580kBTZ3mMh25QZkmsLMRWCLtjeeeedwf/+wx/+MPjf8+fP5+mnnw7tqD5hW8suPmrZEZZzr8taw6lZq0Y95t1332b//r389a9P4vP5uO22L+P19o54vNVq4ZFH/syrr77Ej398D0888SwxMTFcdtmFfOlLNwc1rmeeeYrvfOcuCgoK2bVrBw899AB//etTXHrp5QB8+tOX8OijjwBw7rmbuPHGa7n99v9Go1GxZcs7PPLIn/nLXx6ltHQBd955Dy6Xk1tvvZGFCxeTk5M72qUBuPfeu/jCF77Ixo1nc+DAfu6889s88cSzPPTQA5x99nlcfvmVbNnyLm++OXYqqMXSzuOP/wutVsv3v///+PSnL+H00zditVq57babeOyxf/DlL9/Knj27uOGGm3jllRdHPFdDQz1PP/0SJpOJRx99hMrKCn772z/idDq46qrLuPzyq4iPH37N9ObNb3P0aDmPP/5PHA4HX/zi1QA0NjZQX1/L7373J9RqNffe+z1ef/1VrrnmC1x66eX8+te/pK+vj8suu0KKNSGEEGKSKrqrSY1NGSygQmW0Yg0g3ZgKQKsEj4hpRjpcB6GsbBdnn30eer0evV7PY4/9g6997SsjHr927XoAMjIyKSqaQ1JSMgBmsxmHwx7UNe+66162bn2fd999i4MH9+N2u0c8NikpiZKSuezevRODQU9+fgEpKans3Lmd3l4PL7/8AgAej4eamuoxC7aenh4aGxvZuPFsABYvXoLZbKa+vo4dO7YPznxt3HgWJtPYG0rnzZs/2Ex9587t1NXV8cc/9hebPp+PpqbGMc8xIC+vAJPp+BrflStXo9PpSEpKxmw243I5RyzY9uzZxcaNZ6HVaklKSmLt2v4G4Lm5eXzta3fw4ovPUV9fx8GD+wfv0Q033MSXv3wden0Md931g6DHKYQQQojhtbrayImb+hkug9ZAgt5MuxRsYpqZFgXbqVmrxpwFCyetVsuJ4ZctLc1D0jB9Pt+Q43U63eB/azSak86nUqlQlOMhK36/76RjvvrVm1m5chUrVqxi1ao13HPPnaOOcdOmC3nnnTfR63Wcf/4FAAQCfu66615KS+cD0NnZgdmcMOp5ABTl5IANRQG/349arT4pIGYsMTExg//t9wf41a8eHhyH1WolKSmJioqjg8eMdn9OPBeAXq8f8Xmf1P/48T8P/N0cOXKYu+/+Lldf/XnOOuscNBr14HmcTic9PT24XD3Y7XYSExPH8cqFEEIIcSJfwEe728qytNAshxyvjLh0ifYX0050ZJVHuWXLVrJ58zv4fD48Hg///d9fx2SKp6amCoD33988rvMlJCRSV1eDoig0NzdRWVk55HG73UZDQx033XQra9eexvvvbyEQ6C+iNBoNfr//pHOeccZGysp2s23bx2zYcBYAK1euGUyStFqt3HDDNbS1tY45vrg4E9nZOWzZ0r8M9sCB/XR2dlBcPIfVq08ZXAb50Ucf4nSOLwFo1arVPPvsvwCoqanm+us/R2+vZ8jrGuv+TNTq1afwzjtv4vV6sdvtbNvWHylcVraLFStWcdllV5CXl8/WrR8M3u+f//wnXHHF5/jMZ67g5z//SUjGIYQQQsxWFncHASVAZtz4U8VDIcOYRluPZdQPeIWINtNihi3SNm48iyNHDnHjjdcSCChceeU1zJ+/gPvuu5uXX36BM844c1znW736FF5++XmuueazFBQUsHTp8iGPm80JXHTRpVx33VVotVpWrlyDx+PB7XazfPlK7rvvbpKTk4c8JybGwJIly/D5fBiNRgBuvPFmfv7zn3LddVcRCAS47bbbg9q/BvC9793Lz372Ix599BF0Oj333Xc/Op2Ob3zjm9x77/d54YVnKSmZF9SSyBPdccf/cP/993HDDVejKAp33fUDjMY4FixYxJ/+9HsefvjX3HTTLaPen4k644wzOXz4ENdf/zmSk1MoLCwG4Jxzzud///dbXH/95wAoLV1AS0szb7/9Jk1Njfzwhz+mr8/Pl798PW+//SbnnHNeSMYjhBBCzDYtrjYAsuIyInL9DGMabp8bZ5/rpObaQkQrlRIFHzEM14ettbWOzMyCCI1o+gp3z7B//etJVq8+haKiYsrLj/DTn/6QP/3p8bBdLxpEUx+2mfBzIb1ZQk/uaWjJ/QwtuZ+hN53v6as1b/FyzZv8YuO96DX6sZ8QYgc7yvnt3ke5Y+V/UJJYBEzv+xmt5J6Oz1h92GSGbZbZu3cPDz74s2Efe+CBh0hNHT2xKTc3j7vv/i5qtQq9PoZvf/tO3n77Df72t8eGPf6xx/4x2SGPy2RfnxBCCDHb9Pn7CKAQMwUFVIurjRRDUkSKNTge7d/W0z5YsAkR7aRgm2WWLVsxqSJq3brTWLfutCFfKy2dzznnnD/ZoYXEZF+fEEIIMZs4vE4e2Pl/JBkS+cbKW8N+vdae9ojtXwNINiSiVWtpc0nwiJg+JHRECCGEEGIW8vq9PLzvz1g9nVR0V9PkbAnr9fwBP209FjIjtH8NQK1Skx6bKkmRYlqJ6oItCrbXCRE15OdBCCFEqASUAI8depJ6eyPXzr8CrUrD1ubtYb1mh6cTX8AX0YIN+pdFSi82MZ1EbcGmVmuG7U8mxGzl9/tQq0/u6yeEEEKM17OVL7HXcoDPzr2Y9dmnsCxtMdtbd9Pn7wvbNVtc7QBkRXBJJPQXbNZjxaMQ00HUFmyxsSYcju5hmzgLMdsoSgCHo4vYWIkgFkIIMTnvNnzAuw0fcFbu6ZyVdzoA67NPocfnZq/lQNiu23asYMs0RrZgSzemEVACWN2dER2HEMGK2tARkymBri4LbW2NgCwFC5ZarR5s+ixCIzruqQq93oDJlBDhcQghhJjO9loO8kzFiyxLXcTlcy8a/Pq8pDmkGJL5sGUHqzNXhOXaLT1tJMUkYtAawnL+YGXEDSRFWiIagCJEsKK2YFOpVCQnyw/ReEnfi9CTeyqEEGImqLXX8+eD/yDfnMsXF12DWnV8oZVapWZ99hperH4dS08HacaUkF+/1dUWFQXSidH+sCiygxEiCFG7JFIIIYSIJgesh3mrfkukhyHEhFjdnfxu72OY9fHcuvSLw/ZBW5u1GhUqtraEPnwkoARodUU20n9ArDaWeL0pKpMie/p6Ij0EEYWkYBNCCCGC8Fb9Fl6seo0+CSoQ04zV3cGvy/6AX/Fz27IbMevjhz0uMSaBxanz+bhlJ/6AP6Rj6PJ04w30kWWMbELkgExjetQlRVZ0VfM/799Dg6M50kMRUUYKNiGEEGIMASVAvaMRn+KnUd5MiWmkydnCz3f9Fnefm9uW3TTmDNf6rFOwex0c7DgS0nG0uNoAIh7pPyDdmBZ1M2z7Ow6hoHDAejjSQxFRRgo2IYQQYgytrnZ6/V6gfx+QENNBVXctD+7+HWqVmjtW/QdFCfljPmdRynwS9PEhXxbZ2nMsITIKlkRC/z42V18Pzj5XpIcy6GhX1bH/r4zwSES0kYJNCCGEGEOdvQEAnVpLja0uwqMRYmwHrIf5ddkfiNfH8V8rbyMryJktjVrDqVmrOWA9QnevLWTjaXW1E683EaczhuyckzEQPBItyyJdfT00OprRa/RU2+vwhrEfnph+pGATQgghxlBrrydWa2BRygKZYRNRb3vrbh7Z/xey4tL5r5W3kRKbNK7nr886BQWFj1t2hmxMra62qNm/Bv1LIgFaXdFRsFV0VaGgcFbu6fgCPqpttZEekogiUrAJIYQQY6izN1AQn0dxQgEdni5svdLqQ0Sndxs+4C+HnqQkoYjbV9xCvN407nOkGVOYl1TC1uYdBJTJ9yFVFIUWV3vU7F8DSDEkoVFpomaGrbyrCr1Gz7n5G1Cr1JTLskhxAinYhBBCiFF4/X00uVopMOcN7gGqtcuySBF93qzbzNMVL7AsbTG3LbuR2Ek0qD4taw0dns7BfVWTYfPa8fg9ZEXJ/jXoX/qZZkyNmuCRo12VlCQUYdQZKTTnU94pBZs4LmobZwshhBDRoNHZREAJUGjOI9eUg0alocZWz7K0xZEemhCDAkqA1+veZVHKfG5adC0atWZS51uWthijNpatzduZnzx3UueKtoTIARnGNFpd7RN6bpOzhacrXiQrLoNCcx5F5gJSY5NRqVTjPpet105rTztrs1YDUJpUwmu1b9PT58aoi53Q+ERkVHXXUm2rJSHGjFkfT0KMmQR9PLHa2Al9bwyQgk0IIYQYRa2tf89agTkPvUZHrilb9rGJqNPoaMbtc3NKxopJF2sAOo2OUzJX8kHTxzj7XJh0cRM+10BRFGzwyVTJMKZxwHp4Qj3ntrXsorK7mlp7PVsaPwTApIuj0JxPoTmf+clzg0rlhOPpkKVJJYP//2rtW1R0V8kHQ9NIQAnwp4N/HzasR6fWYtabWZe1hguKzhn3uaVgE0IIIUZRa28gKSaRhBgzAEUJ+Wxt3o4/4A/JG2MhQmFgz9PcY2/6Q2F99ilsbvyQHa17OCvv9Amfp9XVRpzWOKmiLxzSjWn4FT/trg60jG8mq6K7ijkJhXx9+c20uNqotddTY6+n1lbPgY7DvFzzBt9a/TUKzHljnutoVyWx2lhy47OB/t8xOrWO8i4p2KaT8s5KunttfGHBVRSb87F5Hdh77di8DmxeOzW2Ol6ueYM1mStIjU0e17mlYBNCCCFGUWdvoPCEN11F5nw2N35Is6uNvGNvsISItKNdVWTGZZAQEx+yc+aYsig057Ol8UM25Kyb8AcU/YEj6ZNaEhYOA9H+TfZWCvRFQT+vp89Ng6OZCwrPQaPWkBufTW58NqfnrAXA7nXwva0/YWvz9qAKtvKuKuYlzUGt6o+W0Kq1lCQWSfDINPNRyw6M2lhWpy9Dp9GR8Yk9m12ebr730U/Y3PABV8y7ZFznltARIYQQYgROrwurp3PIm67ChAJAgkdE9PAFfFTaaihNmhPyc28qOAuLu4Ntrbsm9HxFUWh1tUXd/jU4XrA1O9rG9bwqWw0KCvNGuN9mfTwr05eys20vXr931HNZ3Z10eDpPOldpUgmtrjZsvfZxjU1ERk9fD3utB1mTuQKdRjfsMUmGRFalL2dry3Z6+tzjOr8UbEIIIcQIBvaqnTjDlmJIIl5nosYm+9hEdKizN+L1e5kXwuWQA5akLqTAnMcrNW/RN4Fmzs4+Fy5fT9TtXwOI0/Uv0xxvwXa0qwqtWkuheeQ9auuyVuPxeyizHBjjXP2zaKWf+Lsb+LPMsk0PO9v24gv4BoNjRnJO/hn0+r182LxtXOeXgk0IIYQYQZ29ARUq8uJzB7+mUqkoTMinRmbYRJQ42lWJChVzE4tDfm6VSsUlxZ+iq7ebD8b5JhNOTIiMnkj/E2UY02gZZ8FW0VVFsblgxJkUgJLEYtJiU9javH3Uc5V3VRKvN5FpHHp/cuOzMWpjpWCbJj5q2UGOKYs8U86ox+XF5zAvcQ6bGz8cV9iNFGxCCCHECGodDWTFZWDQxgz5epE5n/YeK84+V4RGJsRx5V2V5MZnE6czhuX885PnMu9Y1LzH1zuu57YOFGzG6C3Ymu3BF2yuvh4anS3MTRq9OFapVKzNWkNFdzWWno5hj1EUhaNdVZQmlZy0v0+tUjMvaQ7lnZUoihL0+MTUa3a2Uu9oZG3W6qD2aZ6Tv4HuXhu72vcGfQ0p2IQQQohhKIpyUuDIgKJj+9jq7A1TPSwhhvD6+6ix1Y24nypULinehLPPxeZjEfbBau1px6CJITEmIUwjm5ysuAxsvQ46PV1BHV/ZPbB/bezlp6dmrkSFio9bdw77eFtPO3avY8S/u9KkErp6u7G6O4Mam4iMj1p2oFapWZOxIqjjF6aUkmFM552G94MuxqVgE0IIIYZhdXfi6usZNuUtPz4XFSrZxyYirtpWi0/xn7QHKtSKEgpYkrqQt+o309PXE/Tz+hMiM6IuIXLA4tSFAOxu3xfU8RVdVejU2qDSH5MMiSxImcfHLTsJKIGTHi//RP+1Tzq+j60iqLGJqecP+NnRuoclqQuJ15uCeo5apebsvNNpcDRR0V0d3HMmM0ghhBBipqqzDzTMPjlYwKCNIduUKQ20RcQd7apCrVIzJ6Ew7Ne6uHgTHl8vb9ZvCfo5/QmR0bkcEiDdmMqc5AJ2tQW3PO1odxXFCYXo1MF1xlqXtYbuXhuHO08uuo52VZJsSCLFMHxPrnRjGgl6s+xji2IHOo7g6HOyboywkU86JXMVJl0cb9e/F9TxUrAJIYQQw6h1NKBT68geId2uyJxPrb1+2E/OhZgqR7sqKTTnYdAawn6tHFMWqzKWsbnhA2y9jjGPd/X1YPc6ojIh8kSn5a+m3tFIe4911OOcfS6anC3MTQx++enS1IWYdHF81LJjyNcDSoCjx/qvjTT7qFKpKE0u4WhXlfyeiVIft+zErI9nYXLpuJ6n1+jYkLOOAx2HaXO1j3m8FGxCCCHEMOrsDeTH54zYLLgwoQC3z0N7j2WKRyZEP7fPQ52jMSxx/iP5dNF5+BQ/r9e9M+axrcfeiEZr4MiAdXmrAMacZavsrgEY135BrVrLmswV7LMcxOk9HlLU5Gyhx+cecylraVIJzj4Xzc7WoK8ppobD6+RAx2FOyVw5oabyG3LXo1Vreafh/TGPlYJNCCGE+AR/wE+Do2nUfSpFx5ZKVss+NhEhVd01BJRAWBpmjyTdmMa6rNV82PQxHe7RgzoGEyKjfIYtxZjEnIQidrWXjXrc0a4q9GodBebcUY/7pHVZa/Arfna07Rn82sAyx7GKP+nHFr22t+4moATGvRxyQLzexCkZK9nWugvXGPtCpWATQgghPqHJ1UJfwDdsQuSAdGMqsdpYaqUfm4iQ8q5KtGotReaCKb3uBYXnAvBq7VujHtfa045OrSPZkDgFo5qc1RnLaHG1jTqTVdHVv39NG+T+tQE5pizy43PZ2rx9MBWwvKuSDGP6mOmZSYZE0o2pUrBFGUVR+KhlB4Xm/El9IHFO/hn0BXzsaN0z6nFSsAkhhBCfMBDXXzhM4MgAtUpNkTlfkiJFxBw9VkCM1sA5HJIMiZyRu46PW3aOuv+m5VjgiFoV/W83V6QvRa1Ss6utbNjHHV4nza5W5k5wNnNd1hqaXa00OJrwB/xUdtcEPTNamjSXyu7qcTVaFuFV72ikxdXG2gnOrg3IjMtgUcp8trXuGvW46P8JEkIIIaZYrb0Bky6OZEPSqMcVJuTT4mrD4/NM0ciE6Ofsc9HobJ7S5ZAn2lRwNjqNjifL/43dO3wASaurnUxjdC+HHBCvN1GaVMLOtrJhe2NNZP/aiVZnLEen1vJRyw7qHA14/d6g9x6WJpXQ6/dS55C+j9Hio5ad6NRaVmcsm/S5zs47A1efa9RjpGATQgghPqH2WMPssXpHFZnzUVCoszdO0ciE6FfR1d+/KdwNs0cSrzdxRcnFVNlq+cHHD/Be40dDkgzdPg9dvd1kRXGk/yetSl+G1dNJvePkn+fB/Wvx49u/NsCoi2V52hJ2tO3hgPUIAHOTioN67tykYlSoKO+UZZHRoM/fx862MpanLSFWGzvp85UmlYy5rFIKNiGEEOIEPX1u2lztQTXGHdjjViP92MQUO9pViV6jpyB+7O/TcDkt51T+95Q7yIvP4amj/+ZnO/9vcDlxW8+xhMgoDxw50bK0xWhUGnYOsyyyoruKOYlFE0oDHLAuaw1un4d3Gt4n15SNSRcX1PNMujhyTVmyjy1K7LUexO1zT3o55ACVSsUXFlw56jFSsAkhhBAnqO6sR0EZdf/aAKPOSIYxXYJHxJQr76qiZJIFRChkxqVz+/Kb+dKiz9Pda+NnO/+Pp8r/TbWtbvDx6cKoi2VhSim72/cNmS10eJ20uNqYN47+a8OZm1RMiiGZvkDfuGdG5yWXUGOrw+v3TmoMYvK2tewi2ZAU0tltsz5+1MelYBNCCCFOUNlZCxDUDBswGDwy3L4XIcKh091NW0/7mD28popKpWJ1xnK+t/abnJl7Gu83fcwzFS+iVWlINSRHenjjsjp9Gd29Nqq6awe/VtHdv/w02CWMI1Gr1IMR8OP9u1uQPA+f4ueVmrfkd00E9fS5OdJVwcpjITVTZXy5pEIIIcQMV9lRS1psCnE6Y1DHFyXk83HrTqzuTtKMKWEenRBwsO0oELn9ayOJ1cZyxbxLODVrNf86+hxxuriIzwCO15K0RejVOna17x0s0I52VRGj0ZM/wf1rJzoz7zR0Gh0LkueN63nzk+ZyevapvFm/mRiNnguKzp30WMT4Heo4QkAJsCxt8ZReVwo2IYQQ4gSVnbUUmwuDPr4oob8HVo29Tgo2MSUOtJdj1MaSa8qO9FCGlRefzX+tui3Sw5iQGI2eJakL2dO+jyvnXoJGraGiq4o5CaFZfhqrjeXc/I3jfp5KpeJzpZ/BF/DzUs0baNVazis4c9LjEeOz13qQeL1p1B6d4SBLIoUQQohjunttdLq7g9q/NiArLgO9Rk+tBI+IKXKgvZy5SXOmRX+z6WhVxjKcfS6OdlVh63XQ2tMeFbOZapWaaxdcwar0ZTxX9QqbGz6M9JBmlb6Aj0Md5SxNXTjlP3vyky6EEEIcc7xhdvCfnqpVagrj8zjYUU61rVb2l4iwsro7sbg6oqKAmKkWJpdi0BjY2VZGZXcVMPn9a6GiVqm5YeHVLEtbzL8qnueDpo8jPaRZ42hXFR5/L0tTF035taVgE0IIIY6ptTegUanHvdTs9JxTsfXa+fmu3/L9j37C81Wv0uxsDdMoxWx29Fi0e7QEjsxEOo2OZWmLKLMc4FDHUQyaGPJMOZEe1iCNWsONiz7PopT5PFn+b7a17Ir0kGaFfZYD6DX6iPzsyR42IYQQ4pgGRxP5CTnoNLpxPW9VxnIWpcxnr+UgO9vKeKt+C2/UvUuOKYvVGcs5JXMliTEJYRq1mE3KuypJMJjJNE6fuPzpaHXGcra17mJ7224WJM+LuvAUrVrLzYuv4+F9f+Zvh/+JVq1hVcbySA9rxgooAfZbD7EouXTc/z6EgsywCSGEEMe0uNrIS5hYkINBa+DUrFV8dflN/Oi0O7ly3qXo1Xqer3qVB3b+ZkhfJyEmos/fx+GOoyzJmI9KpYr0cGa00qQSTLo4Akogapef6jQ6bln6RYoTCnns0JMyqx9GdfZGbF4HS9OmfjkkSMEmhBBCAP39dbp7beQmZE36XPF6E2fmnsY3V3+Va+dfQVdvt7yZEpO2u30fLl8PZxaujfRQZjyNWsPy9CUAzE2Mjv1rw4nR6Ll+4ecIKAFqjjUrF6G3z3oQtUrN4pT5Ebm+LIkUQggh6J9dA8if4AzbSEqT5gJQaashNz46Y9jF9PB+08ekx6ayOKOUDqsr0sOZ8c7PP4tEvZm8+OjZvzacZEMiOrWWNrcl0kOZsfZZDjI3sRhjkP05Q01m2IQQQgigxdU/A5Yb4oItJTaJpJhEqrprQnpeMbs0Opqpsddxes5aifOfIimxSVxQdG7U32+1Sk1abCrtPdZID2VGanO109rTHrHlkCAFmxBCCAH0z7DpNXpSjUkhP/ecxEIqu2sk8l9M2PtNH6FTa1mbtTrSQxFRKN0oBVu47LMeAmBp6sKIjUEKNiGEEAJodrWRFZcRlk/TSxKLsHsdWNwdIT+3mPncPg/b2/awMn0ZcRFakiWiW7oxDau7Q8KNwmCv5SB58TkkG0L/YV6wpGATQggh6F8SmRWXEZZzz0koApBlkWJCdrTuxuv3ckbOukgPRUSp9NhU/IqfTk9XpIcyo9h6HdTa61kWgWbZJ5KCTQghxKzn9LpweJ1kx2WG5fyZcenE6YxU2oIv2FpdbbxT/15YxiOmD0VReL/pY/JM2RSa8yI9HBGl0oypALTJssiQOmA9hIIS0f1rIAWbEEIIMRg4Eq4ZNrVKzZyEonHNsL1a+zbPVL6Ex+cJy5jE9FBlq6XZ1coZOeuk95oYUYYxDYD2HkmKDKW91oOkGJLD9mFesKRgE0IIMesNRPqHq2CD/uARi7sDW699zGP7Aj4OWA8D0B3E8WLmer/pIwwaA6szV0R6KCKKmXRxxGoNEjwSQh6fh/KuSpalLYr4hyVSsAkhhJj1ml1txGoNJMYkhO0aJYn9+9gqg5hlK++swOPvBaC71xa2MYno5vA6KWvfz6lZK4nR6CM9HBHFVCoV6bFp03aGTVEU/njgcfZZDkZ6KIMOdR7FF/BFNB1ygBRsQgghZr2BwJFwfoqaZ8pBr9ZRFcQ+tjLLAVT0jyWYGblQUBRFisMo81HLDnyKn9Oz10Z6KGIaSDemYnFPzxk2u9fJnvZ97GwrG/dz36l/j/LOypCPaZ/lIHE6I8UJhSE/93hJwSaEEGJWUxSFlmOR/uGkUWsoSigYc4bNH/Czz3pwcJP7VBVRZZYD3LX1x3S4JWUuGgSUAB80baMksYhsU2T3z8xkiqLg88+MKPw0Yyqdnm76/H2RHsq4tR5blt7gaBrX8/r8ffy76hX+evgpvCF83f6AnwMdR1iSshCNWhOy806UFGxCCCFmNbvXiauvh6wp2FQ+J7GIZmcrPX3uEY+p7K7B1dfDKRkriNXGYvNOzQzbka4KAkqAekfjlFxPjO5wZwUdnk6J8g+zrQdauf7u1/B4fZEeyqRlxKaioEzLfo8D+4jb3Vbc4whaanK1EFACdPfa2Nz4QcjGU9Fdjdvnjng65AAp2IQQQsxq4U6IPNHcxCIUFKpttSMeU2Y5gE6tY0FKKYkx5ikLHamx1QHQ5GyZkuuJ0b3f9BHxOhPL0xZHeigz2v7qDhw9fTS2uyI9lElLH0iKnIbLIlt62gb/ezy/g+rt/R8w5cfn8Ebdu7j6ekIynn3Wg/2/h5PnhuR8kyUFmxBCiFnteEJk+GfYCs35aFQaqkYo2AJKgL2WAyxKKSVGoycxJmFKlkR6fB6anf2Fa/OxAlZETqeniwPWw6zLXoNWrY30cGa06ub+D0Tq2hwRHsnkDfRim47BI62uNlINycD4lkXWO5ow6eK4bsHn8Ph6eb32nUmPRVEU9lsPMz95LvooCfuRgk0IIcSs1uJqJU5nxKw3hf1aeo2e/PgcKrurh3281t6AzWtn2bFZlYQY85SEjtTaG1BQMOniZIYtCnzYvB2A07NPjfBIZjaby4vV1r/8rn4GFGyxWgNmfTyWaRjt3+pqpzS5BLM+fpwFWyP58blkmzI5NXMVWxo/nPQ+3Naedjo9XSxKmT+p84SSFGxCCCFmtYHAkanqszMnsYg6e+OwG+TLLPvRqDQsTlkAQGJMAnavg4AS3lCEGlsdKlScmrmKDncnHl9vWK8nTmb3OtjSuJVf7Potr9W+zeLU+aTEJkd6WDNadXP/7LUpVkd9uzPCowmNtNhU2qZZwebwOnH2uciMyyAvPifogs3r76PF1UZ+fA4AFxWfj0ql4qWa1yc1noMdRwBYlFI6qfOEUlAF24svvsiFF17I+eefz9///veTHj948CCf/exnueSSS7jllluw26XJpxBCiOinKArNzjayp2A55ICSxCL8ip86e/1JY9nbfoDSpBKMulgAEmPMBJQAdm94P/2vtteRGZfOnMRCFJTBZaIivFx9PWxt3s6v9/yB//3gh/zz6HO4fR4uLt7EFxZcFenhzXjVzXY0ahUbV+bSZHHOiLTIDGMq7e7ptSRycFm6sb9ga+1pDyrxscnZTEAJkG/OBSDJkMiZuaezo3UPjY7mCY/nUEc5mXEZJBuSJnyOUBuzYGtra+PBBx/kH//4B8899xxPPfUUlZVDex3cd9993H777bzwwgsUFRXx6KOPhm3AQgghRKh099rw+D1TEjgyYKCnT2V37ZCvNzpbsHo6WZ5+PGQiQW8GwtuLLaAEqLXVU5xQQI4pC4BmlyyLDLcny//N//vgXv5+5Gk6PJ1sKjyb757yX3z31P/iU4XnYNLFRXqIM151s53cNBPzC5Px+RVaOkITWBFJ6cY0HF4nbt/ISbTRZiDSP8vUX7AFlEBQv4Pqj83E5cfnDn7t/IIzidUaeL7q1QmNxePrpaq7hkXJ0TO7BkEUbFu3bmXt2rUkJiZiNBrZtGkTr7322pBjAoEALld/uo7b7cZgMIRntEIIIUQIHQ8cmbqCLU5nJDsu86QG2nst+1GhYmnq8RjpxJgEgLAmRbb3WOjxuSkyF5BsSEKv0dPklOCRcPIFfHzYvI15SXP49urb+f7a/+Hi4k3Sb20KBQIKNS12inPMzMnp/zmbCfvY0geDR6bPssgWVxsGjYEEvZk8UzYQXPBIvb2ReJ1p8PckgFFnZFPh2RzqLJ9QM+2jXZX4FD8Lo2g5JARRsLW3t5OWljb45/T0dNrahi6V+M53vsOdd97J6aefztatW7n66qtDP1IhhBAixJoHI/2n9o1ySWIR1bZa/AH/4NfKLAcoSSwi/oTwk4Rjb0RsYUyKrLb1L80sTihArVKTE5dJswSPhFWHu5OAEmBNxgryzblTtn9SHNfc4cLj9VOcZSY7zYReq6a+bfrvY0uLnZ4FW1ZcOiqVimRDEkZtbHAFm6ORPHPOST8/G3PWkxSTyHNVL497/+/BznL0Gj1zEovG9bxwGzMrNhAIDLkRiqIM+bPH4+G73/0ujz32GEuXLuXPf/4z3/72t/n9738f9CBSUsKfzDWbpKXFR3oIM47c09CS+xl6ck8npqumkwSDmaKcoQVbuO/nCvcC3mv6CJfOxpzkAprtrbS42vjiiiuHXDtFiUOjUuPVeMI2ppaaJkz6OBYVFKNSqShOzWdb4x5SU00hKyTk+3OoOm//7GppTgFpKRO7N3JPJ2dPdScAa5Zko1GrKMpOoKWrZ9rf14RkA6rtKlwqe8RfS7DXb3dbWJm9ZPD4OSn5tLhbR31+r89LS08b6wpWDnvc55ddym+2/4VKz1FOy18T1DgURaH846MszZhPdkb07F+DIAq2zMxMdu7cOfhni8VCenr64J+PHj1KTEwMS5cuBeBzn/scDz300LgG0dHhJBBQxvUcMby0tHgsluk/pR9N5J6GltzP0JN7OnE1HY1kxKYPuX9TcT/T1f17xXbWHMDsT+ad2m0AlMTOPenaZr2Z5i5L2MZ0qL2Kgvg8rNb+2YVkbQpOr4vKpqYhS40mSr4/T1bR2j+rqes1TujeyD2dvL3lbcQZtOiOzcBkpRjZdqiN9nb7tJ/xTDYkUmNtiuj3SLDfo06vC1uvgyRN8uDxGTGZbG7/gNa2bjRqzbDPq7bVoSgKKZq0Ya8zP24BOaYs/r7nOYpjSoLqZ9jqasPS08k5eRun/N6p1apRJ7DGXBK5fv16PvroIzo7O3G73bzxxhts2LBh8PGCggJaW1upru7vKfP222+zZMmSEAxdCCGECJ+AEqDF1Ub2FO5fG5AYk0CKIZnKYw20yywHKDDnkWRIHObY8PVi6+lz0+pqozihYPBrOceWh8o+tvBp77Fi0sURpzNGeiizVnWznaJs82Bxlp9uwt3rG+zLNp2lG9OmzZLIgX3EmSf8Hs6Lz8Gn+EdNq623NwJQYM4d9nG1Ss2lcy7E6unkg6ZtQY3lYEc5AAuTo6f/2oAxC7aMjAzuuOMOrr/+ei677DIuuugili5dys0338z+/ftJSEjgxz/+Md/4xje4+OKLeeaZZ/jRj340FWMXQgghJqzL043X753SwJETlSQWUdVdQ4e7i3pHI8vTFg97XEJMAt1h2sNWc6y1QJH5eMGWPZAUKfvYwqa9xzIYDiGmnrvXR5PFRXGWefBr+Rn9y+pmSvBIe48VRYn+1WutPQPBT8dX7wUTPFLvaCRebxpM0h3OwuR5lCQW8Wb9ZnwB35hjOdRRTqYxnZTY6FoOCUEsiQS4+OKLufjii4d87Q9/+MPgf2/cuJGNGzeGdmRCCCFEGEUqcGRASWIR21p38Ub9uwCjFGxmjnRWhGUMAw2zC8x5g1+L0xlJjEmQGbYwau+xsiB5XqSHMWvVtjpQYDAdEiA3LQ6VCurbnKwqTR/5ydNAWmwqHr8HR58Tsz669+S1uNqJ0ehJikkc/FqaMZUYjZ4GZxPrGH7/Wb2jkYL40QN7VCoVmwrO5jd7H2VH6x7WZY+8l83j66Wyu5oNuesn/FrCKajG2UIIIcRME4lI/xMNpJB92LSN7LhM0o1pwx6XGGPG4/fg8fWGfAw1tjpyTFkYtDFDvp5typRebGHi8fVi89plhi2Cqpv7Z6yLTphh0+s0ZKXEzZAZtv7fJdNhWWSrq43MuIwhhZdapSbXlD3iDFuv30urq528+OGXQ55oQfI8ck3ZvFm/edTEyIruKnyKn0Up0bccEqRgE0IIMUu1uNpIjEnAqIuNyPXTY1OJ15lQUEacXYPjvdhs3tDuYwsoAWrt9RSdsH9tQE5cFq2u9iFtB0RoWNz9b6JHKtBF+FU328lINmKK1Q35en6Gifr26R/tnzGNerG1uNrIMp78oVlefA6NjuZhi6xGRzMKyoj7106kUqk4v+BM2nos7LMeGvG4gx2RjfP3B0ZvPyAFmxBCiFmpxdkasdk16H8jMfDmYHn6yGFdiTH9swCh7sXW4mrD4++lyJx/0mPZpkz8ip+2HktIryn6968BMsMWIYqiUNVsH7J/bUB+ejxdjl7sPd4IjCx0kg1JaFSawe+1aOXq68HudZAZd/IS1Lz4HLyBvmGLznpH4+AxwVietoTU2BTeqH132H19iqJwqOMIpUkl6IJIkwyHf79XM+rjUrAJIYSYdQJKgNae9ogWbAAbc9exMfc0skfZRzfQPLs7xEmR1bY6AIoTCk96LEeCR8Jm4A1oWmxKhEcyO3XYPdhdXubkDFOwZfTHqjdM8wbaapWatNgU2t3RPcPW6moHhl+WPlCMDbcsst7RSII+Pui2Ixq1hnPzN1LnaOBoV9VJj7f1WOjwdLEopXQ8ww+p8obuUR+Xgk0IIcSsY3V30BfwRSxwZMC8pBKumnfpqBvnEwcLttDOsNXY6jDp4kiNTT7psQxjGmqVmiaXBI+EWluPlaSYRPQafaSHMitVN/d/8FGcPVzBNnOSItOMqVE/w9YyGPx0csGWaUxHq9aOULA1BbV/7URrM1dh1sfzRt27Jz12qOMIELk4/y5HLzbn6HuUpWATQggx6wwEjmSbIjvDFowYjZ5YrSHkM2w19jqKEwqHLRa1ai2ZxnSZYQuDdrdE+kdSdbMdnVZNbtrJTYpNsTpSzDEzYh9bujEVi7tj1KCNSGt1taNX64btP6lRa8iJyzqpYPP4emlztZMfxP61E+k0Os7OO4MjXRWDPdwGHIxwnH9ty9i/26VgE0IIMesMNmsdZrN7NErQm0O6h83pddHeY6Uo4eT9awOyTZkS7R9iiqLQ3mOVwJEIqmq2UZAZj1Yz/FvgvPT4GTHDlhGbhi/go8sTnh6OodDiaiMzLh21aoS/i/hsGpzNQ/adNTr7A0fyg9y/dqLTc9YSqzXwRv3mwa/1+r1UdlezMILLIatb7GjUI6+yACnYhBBCzELNzlaSDUknxdlHq8SYhJDOsNXY+/evndgw+5Ny4rLo6u2mp88dsuvOds4+F26fW2bYIsTnD1DX6mTOMMshB+RnmGjt6KHXO70TUge+x9rd0bsssn8f8cjL0vPic3D73HR4uga/NhA4kj/OJZEAsVoDZ+Sso6x9/+By0aNdlRGP869utpOZHDfqMVKwCSGEmHVaXG1kRzhwZDwSYszYQliwVdvqUKvUo8ZiZ5v630g1yz62kBkIHEmPlYItEhranfj8AYqzRw6rKMiIRwEaLdN7WWS092Lr6XPT3WsbNiFywHDBI/X2JhL0ZhJiRi66R3NW3ulo1Breqt8CwKEIx/kHFIXaVjt5GScv0T2RFGxCCCFmFX/AT3uPJeKBI+ORGJOAzWsP2X6UGlsduabsUYMvJCky9I5H+suSyEgYCBwZbYZt4I3zdF8WadbHE6PRR23wSGvPyAmRA7LjMlGr1EMLNkfjuPevncisj2dd1hq2teyiu9fGwY5ySpPmRCzOv7WjB3evn/x0KdiEEEKIQRa3FZ/ij3ik/3gkxpgJKAEcXtekz+UP+KmzNwzbMHvoNROI1cZKUmQItbutqFVqUgyRCTeY7aqabSSY9CTFj7wUOsVsIM6gpW6aR/urVCrSY1Ojdoat9dg+4tF+D+s0OrLiMgYLNo/PQ3uPZUL71050bv4G/EqAJ8v/TYenM2LpkAA1xwJH8o4llI5ECjYhhBCzSvPAG4VpkBA5YKAXWyiCR5pcLXgDfRSPUbCpVCqy4zJlhi2E2nsspMWmoFFrIj2UWam62c6c7IRR22ioVCryM+JpaJ/eM2zQP5Mbrb3YWlxt6NQ6ksf48CLPlEODowlFUWhwDASOTHyGDSA1NoVVGcvYbz0EENH+a9XNdgx6DWkJhlGPk4JNCCHErNLibEWFikzjyHsnok3isf0aoejFVmOrB0YPHBmQY8qi2dk6JKVNTFx/QqTsX4sER4+X9i73sP3XPikv3USjxYU/EL2R+MFIN6bS4e7EF/BFeignaXG1kXms3+No8uJzcPQ5sXntNAwEjkxiSeSA8wvOAiDDmE7KML0op0p1i52iLDNqSYkUQgghjmvpaSclNnlaNS5OGCzYJh88UmOrI0EfT/IwvY8+KduUicffS+cJKW1iYgJKgHa3lfRY2b8WCQNLz0bbvzagICOePl+A1o6ecA8rrNJiU1FQsLo7Iz2Uk7S62skMYln6icEjdY5GEmMSMOtHXz4YjBxTFpsKzub8gjMnfa6J6vP5aWx3UpQ19vekFGxCCCFmFes0nOUw6+NRq9QhWRJZbaujaISG2Z80GDwi+9gmrctjwxfwTbvvvZmiqsmOSgUFmWO/2c8fDB6Z3vvYjidFRlfwiNvnoau3O6h9xDmmLFSoaHA00eBomvRyyBNdMudTrM1aHbLzjVddmxN/QAlq1lcKNiGEELOGoihY3J2kGlIiPZRxUavUmPXxdHsnN8Nm63XQ4ekctWH2iQZaHzTJPrZJsxzbSyQJkZFR3WInN82EQT92GmBmihGdVk3dNE+KPN6LLbr2sbW6+hMig5lhM2hjSDemUtFVTVuPJaQFW6TVHEstlRk2IYQQ4gQ9Pjcev4fUCO5ZmKhQ9GKrPdYwuzihMKjjDVoDKYZkmp0ywzZZxyP9ZYZtqgUUhepme1AzGQAatZrctDga2qf3DFuczohJFxd1SZHHEyKD20ecF5/D0e4qAPLNk0uIjCY1LXaS4mNGTS0dIAWbEEKIWcPq7gD6U8Kmm8SYhEmHjgy8cRtPS4NsU6bMsIVAe48VvUZPgn5iDX/FxLV19uDu9QVdsAHkZ8RT3+aY9oE76cbUqFsS2dLThlatDfr3cN4JMf4zaYatutlOcRCzayAFmxBCiFnkeME2/WbYEmPMkw4dsXntxGj0xGpHj5A+UY4pi3a3lT5/36SuPdu1uS1kxKYGtXdQhFZ5QzcAxdkJQT8nP92Ey+Ojw+4J06imRloU9mJrdbWTEURC5IA8U3/BlhSTSLx+9AbT04XT3Ud7t5uiID9EkIJNCCHErDGQlpZimH4FW4LejNvnxuv3Tvgc3b32wcTJYGXHZRJQArT2tE/4umIg0l/2r0212lY7/3ynkpy0OLJSjEE/L/9YI+OGGRA8YvPa8fh6Iz2UQS2utnHN8ufGZwOhifOPFtXH9q/JDJsQQgjxCVZ3J/E6Ewbt2HsGok3isebZk1kWaeu1k6gPfpYBTkiKlH1sE+YL+Ohwd8r+tSnW0uHiF0/tJc6g444rl6Eex+xmbroJlYoZEzxiiZLgEY+vv01IpjH4gi1OZ+SsvNM5LfuUMI5satW02FERXGopSMEmhBBiFrF6OqflckgITS82W69t3DNsabEpaNVa2cc2CVZ3JwqKzLBNIavNzQNPlqFWq/jmNctJNge/DBggRqchM9k47aP9s+MyAThgPRLhkfRrOzZTn2UKvmADuGLuJSxKmR+OIUVETYud7NQ4YmPGTi0FKdiEEELMIh3uDlKmacE22Rk2RVH6Z9hixjfDplFryIrLkF5skyAJkVPL5vLywJNl9Hr9/PfnlpORFPxSyBPlZ8TT0D69Z9gy49JZlrqIN+rfnXRoUSi0DCREGoNLiJyJlGOppcHuXwMp2IQQQswSvoCPTk83adMwIRL6Q0eACUf7u/p68Cn+cc+wQf+n9DLDNnEDfbDSY6VgC7ceTx+/eKqMbmcv37hqGXnpEw+pyM8w0WHvxeaMnv1fE3H53IsIKAGeq3wl0kOh1dWORqWZlkm9oWKxeXC6+4LevwZSsAkhhJglOj3dKCikTNM3CgatAYMmZsIFm+1Y0+2JFGw5pizsXgcO7/ReHhYp7T0WTLo4jLqJzfTMZoqi0NrZw/v7mnns1cP837P7eeXjOo42dNPn8w85ttfr55f/2kez1cXXL19KSc74ZpM/aUlR/++KPRXRsf9rolJjUzgnbwM72vZQbasN+/WcfS56vO5hH2txtZFhTEOj1oR9HNFqPA2zBwS3cFIIIYSY5jqOJUSmTsOEyAEJk+jFNvC8xAkUbJnHGty291hnTKz2VJKEyOAFFIWqJhuVjTYqGm1UNtlwuvtbShhjtJiMOnYf7V9iqtWoKMiMZ25OIiW5CWze00RVs43/uHQxi4om/3OekxZHRrKRneXtnLliejdsPr/gLD5u2cm/jj7Pt1Z/PehI/fFyeJ1876Of4PV7idUaSDYkkWxIPPb/STQ4GilJLA7LtaeL6mY7eq2anLS4oJ8jBZsQQohZweqZvj3YBiRMohfbwMzcRBo3JxuSAOjq7Z7QtWe79h4LC1JKIz2MaeHf71Xz8kd1AGQkxbJsTgoluQmU5CaSlWJErVJh7/FS1Wij4lhh99auBl7bXg/Aly6Yz+r5odkfpVKpWF2axqsf1+Po8RJv1IfkvJFg0MZwWcmF/OXQk3zcspP1YUpc/LhlJ16/lysWXUh7dzedni463F1UdNXg8ff3tDuxEfZsVNNiJz8zHq0m+KJZCjYhhBCzgtXdiVatndCSwGiRGGOmoqt6Qs8dmGEzT+D1Jx0LKun0dE3o2rOZx+fB5nWQESszbGNRFIWtB1pZVJjEzRcvwhw3fIFkNupZMS+NFfP672mfz09tqwO1WsWccTTHDsbq0nRe/qiOPRVWNizLDum5p9qajBW83/QRL1S9xor0JcRqY0N6fkVR2NqyneKEQq5afDEWy9DAlp4+N3avnbRZvJfT5w9Q1+bgrHHO2MoeNiGEELOC1d1JiiE5bEuBpkJiTAI2r52AEhj3c229dky6OHTq8X9Wa9AaMGpj6fJ0j/u5s91A4EiaJESOqb7NSZejl1MXZo5YrA1Hp9UwNzcx5MUa9AePpCYY2FVuCfm5p5pKpeLKuZfi7HPxSs1bIT9/ZXc17T3WEfulGXWxZMZlzOr9a00WF32+AMXjSIgEKdiEEELMElZ3x7ReDgn9SyIDSgBnn2vcz+3utU9qdjHJkEinFGzj1t5zLCFSCrYx7amwoFLB0pLoCQZSqVSsnp/OodpOXJ6+SA9n0vLNuazLWs3mxg9pdbWH9NwfNG8jVmtgZfrSkJ53JqluGX/gCEjBJoQQYhZQFAWre/o2zR4w0ENtIkmRNu/kCrZkQ5IsiZyAgR5ss3kZWLDKKqyU5CRgjrK9YqtK0/AHFMqmeVrkgEvmXIBereeZihdRFCUk53T19VBmOcCajJXoNdH19xdNqpttxBt1pCaMr5G7FGxCCCFmPJevB4/fM60TIuF4wuNEkiK7e20k6ie+ZCzZkCihIxPQ3mMlKSYRvUYX6aFEtQ6bh/p2J8vnRl9hW5xlJtkcMyOWRQLE601cWHQuhzrLOdBxOCTn3N66G1/AN+JySNGvpsVBUZYZlUo1rudJwSaEEGLGG4j0n6492AYMzLCNNynSH/Dj9LomtyQyJhG3z4PbN3x/JTG89h4rGRLpP6ayyv7Zq+Ul0VewqVQqVs1L50BNB+5eX6SHExIbc9eTYUzjmYoX6QtM7jUpisKHzdsoiM8jN356B7OEk7vXR4vVNe79ayAFmxBCiFnA6p7+kf4A8ToTKlTYxjnDZvc6UFAm1INtwEC0v+xjC56iKLS7LbJ/LQhlFRYyk41kpQTfm2oqrZ6fhs+vsLdqZiyL1Kq1XF5yERZ3B3vb90/qXDX2elpcbTK7NobaFjsK/TO24yUFmxBCiBnPMtA0e5rPsGnUGsx607hn2AaOn9wetkQASYocB2efC7fPI02zx9Dj8XGkvjsql0MOmJOTQIJJz64jM2NZJMCC5HloVBoanS2TOs+HzdvQa/SsylgWopHNTAOBI4VSsAkhhBAn63B3EK83ETMDNsMnxCSMew+bzdv/RmFgSeVEJB0r2GSGLXhtxwJHZIZtdAdqOvAHlKhcDjlArVKxal4a+6o78HhnxrJIjVpDujGVFlfbhM/h9nnY3baX1enLMWjHF6Qx21Q02khPisUUO/79rFKwCSGEmPGs7k5SDdN7dm1AYkzCuFMiBwq8ycywmfXxaFQaCR4Zh8FIf2maPaqyCiumWB0lOaHvoxZKq0vT6fMF2F/dGemhhExWXMakCradbXvwBvo4LUeWQ46m29nLgepOVs6b2O8CKdiEEELMeFbP9I/0H5AYYx53wWbrtaNWqTHpJr4/SK1SkxSTINH+49DeY0Gj0gwuJxUn8/kD7KvqYFlJCmr1+JLzptq8vETijTp2Hglt/7JIyorLoMPTSa/fO6Hnf9i8nRxTFgXxeSEe2czywb4WAorChmUTC2XRhng8QgghRFTxBXx0ebpJzZwZBVtCTAIuXw9ef1/QUfG2XjsJejNq1eQ+p5Xm2UN5fB7+sP9vqFQqCs35FCXkU2jOJ05nBKDdbSU1NgWNWhPhkUavikYbPb0+lpdE/yykWt2/LPKjg214+/zoddP/7zUrLhOAVlcbBebxFV31jkYaHE1cOe/SccfUzyYBReG9vc3Mz08kM9k4oXNIwSaEEGJG6/R0o6BM+0j/AQNJj7ZeO2nG4F6Trdc+qYTIAcmGJI52VU36PDOBoig8fvhflHdVkm3K5LXat1Hob0KcYUyj0JxPra2efHNuhEca3fZUWNBq1Cwumh4fqKwqTWdzWTMHaia+vC2aZMdlANAygYLtw+bt6NRaTslYEY6hzRiHa7uw2jx8duOcCZ9DCjYhhBAz2kCkf9oMKdgSTmieHWzB1t1rIzMufdLXTjIk0t1rwx/wz/pZo3ca3mePZT+XzbmQ8wrOxOPrpd7RSK2tnmp7HYc6ynH0OcmLz4n0UKOWoiiUVVhZWJhEjH56fD+V5icSZ9Cys7x9RhRsqbEpaFWace9j6/V72dm6hxXpSzHqJjZrNFtsKWvCFKub1PeLFGxCCCFmNOtgpP/0+AR/LANJj+PpxWbz2ilNLpn0tZMNiSgodPfaSYlNmvT5okmzs5VkQxIGbcyYx1Z0VfNc1SssS1vMufkbATBoY5iXNId5Sf2foiuKgs1rJ15nCuu4p7MmqwurzcOF6woiPZSgaTVqVsxLY+eRdvp8AXTa6R0HoVFryIhLH3fBtrttLx5/L6dlnxqmkc0MNpeXPRVWzl2dO6nvlen9XSaEEEKMwerpQKvWYtbHR3ooITGwtLHbG1zwSK/fi9vnIVE/+QS+5Jj+Im2mJUU2OVv40fYHuX/nr8Z842rrtfOng38n1ZDMdQuuHHHvjkqlIjEmYdbPRI5mT0V/imY0x/kPZ3VpOh6vn4O1MyMtciJJkR82byfDmM6chMLwDGqG+HB/C/7AxMNGBkjBJoQQYkbrcHeSYkiedOBGtDBoDOg1+qB7sdlCEOk/4HgvtpmVFPlG3bvoNTp6+tzcv/PX7G7fN+xx/oCfRw88jsfn4eYl1xOrjZ3ikc4sZRVWirLMJJrGntWMJgsLk4iN0bKrfHxpkQFF4c2dDfzgsR10O3vDNLrxy4rLoNPThcfnCer4Lk83NfY61mWtlrCRUQyEjczLSyQrZeIJvSAFmxBCiBnO6p45kf4wMHNjpjvIaP+B40JRsCXPwObZ7T1WdrXtZUPOer5zyn+SE5fFowce59mKl/AH/EOOfa7qFapstXx+/hVkmzIjNOKZodvZS02LneVzp9fsGvQvi1xeksqeo1baunqCek63s5df/nMvT7xVQW2rg+2Ho6c1QNZg8EhwY6q21QFQmjT5ZdYzWXldF+1dbjYun9zsGkjBJoQQYgZTFGXGFWwAifrgm2cPHBeKlEi9Ro9JF0fXDJphe7NuMxq1hrPyziAxJoFvrLyFDTnrebvhPX5d9gfsXgcAu9r28k7D+2zMXc+aTEnFm6y9lf3LIVdMw4IN4KyVOfgCAb77+2089uphOu0jz07tKrfwvUe3c7Shm+s2lZKbFsfucc7OhVPWCUmRwaix16FT68gxZYVzWNPelr3NxBm0rC6dfDiNhI4IIYSYsVx9PXj8HlJnSELkgISYBKptNUEd2z24JHLye9jgWC+2GbKHrcvTzbbWXZyWfSoJMf17HLVqLZ8rvYxCcx5PlD/LT7Y/xKVzLuCpo/+myFzA5SUXRXjUM8OeCiupCQZyUie3VCxSSnIS+Okt63j5ozo2lzWx9UArZy7P4dPrC0mI0wPg8fp44q0K3t/XQkFGPF+5ZCFZKXHYXV5e+KAGm8s7eGwkpcamoFNraXG1BnV8ra2e/Phc2Z85CnuPl13lFs5emYtOO/n7JAWbEEKIGcvq6Y/0TzXMsBm2GDO2XjuKooy5h8TmtaPX6DFoQrNPKDkmkTa3NSTnirS3699DQRlMejzRqVmryDFl8Yf9f+Wvh5/CpIvjpsXXolXLW6fJ6vX6OVTbxZkrsqf1HqgEUwyfP28em07J58WtNbyzu4n39jVz7qo8FhQk8bfXy7F0u/n0ugIuPb0IraZ/YduqeWk8/0ENeyosnLk88m0f1Co1GcbgkiL7Aj4aHE2cmXf6FIxs+tq6v7U/bCQEyyFBCjYhhBAz2PFI/5k1w5YSm4RP8dPp6R4zXr/7WNPsUL0xTjYkcaSrIqhiMZo5vE4+aN7GKRkrR7yHufHZfHvN7bxS+xar0pcNhq6IyTlY24nPH2DF3OnfxwwgJcHAFy9YwAWnFvD8BzW8+nEdr3xcR4rZwLevXcm8vMQhx+ekxZGRFMuu8ugo2ACy4jKp6K4a87hGRxM+xU+ROX8KRjU9KYrClr3NzM1NCNkMshRsQgghZqyBgi1lhu1hyzX1f2rb6Gwas2Cz9dpJ0E9+/9qAJEPisVYB7mndMPedhvfxBXycX3DmqMcZdUaumHvJ1AxqFrDa3Dz3fjVxBi1zc0OzTDdaZCQb+coli7hwbQH7azrYuCwHo+Hkt9oqlYqVpWm8sb0Bl6ePOIMuAqMdKjsugx1tu3H73KOmn9YcCxwpTJCCbSRHG7pp6+zhonULQnZOCR0RQggxY3W4O4jXm4jRRH6fSCjlmLJQoaLB0TzmsbZe+2Cz7VBImgFJkT19bt5r/Ijl6UvIiEuP9HBmjfL6Lu79y0467L3ccumiwSWCM01uuokLTi0YtlgbsGpeOv6AMhi+EmlZpuCSImvs9STFJIb0d8pMs6WsGWOMljXzQ/e7ZWb+pAghhBCAxd1B2gxbDgn9aY0ZxjQanU2jHqcoCjavPSSR/gNSDP0zetO5F9t7TVvx+D1sKjg70kOZNd7d08QDT5YRZ9Bx5/WrWFw0834ux6MoK56k+Bh2lVsiPRTgxKTI0YNHamz1FCcUTMWQpiWnu4+d5e2sW5yJXhe6UBYp2IQQQsxYVncnKYaZ+cYwNz57zBk2l68HX8AXnhm2aZoU2ev38m7DByxKmU9efGgCAcTIfP4Af329nL+9Xs6iomTuvH71pJsIzwQqlYpV89I4UNOJx+uL9HBINiShV+tocY4cPNLda6Ort1uWQ45i64FWfH6FjctC+7tFCjYhhBAzki/go7vXNuN6sA3Ii8+hu9eG0+sa8RhbCJtmDzDp4tCqtXRN0yWRHzZvw9nn4lOFMrsWbvYeLw88WcbmPU1csDaf2z+7dNRlgrPNqtI0+nwBDlR3jnmstdvNr57eR5ejNyxjUavUZMaNnhRZY6sHoMgsM2wj2XqgheJsM7npppCeVwo2IYQQM1KnpwsFZcYWbAPBIw2jLIvsDmHT7AFqlZqkmIRpuSSyL+Dj7fr3mJtYTHFCYaSHM6M1Wpzc+9gOalrsfOXihVx5Zglq9fRNFQ2HubmJxBt17Do69rLIv795lLJKKx/sbwnbeLLiMkddElljr0Or0pArM9PD6rR7qG9zsmpe6NNPpWATQggxRHuPBbfPE+lhjOm12rd5seq1ER+fqZH+A/Li++PAG0dZFmkbaJodwpRI6F8+NR1n2La37KK718YmmV0Lu7+8eoQ+v8J3rl3J2kWZkR5OVFKrVayYm8beSit9vsCIx5VVWtlb1YFWo2b74bF7pU1UVlwGNq+Dnr6eYR+vsdWTF5+LTnoRDmtvVX/fz2UlqSE/txRsQgghBrl9bn68/Zf8bOf/0X3szX60+qBpG6/VvcMB6+FhHz9esM3MGbY4nZGkmEQaHCPPsA0siTSHcIYN+vexTbeUyIAS4I36zeTH5zI/aW6khzOjNbQ7qWq2c+Gp+RRlhfZ7b6ZZVZqGx+vnUO3wyyL7fH6eeOsoWSlGrthYTJPFRZPFGZaxDASPNA+zLNIX8NHgaKRI9q+NaG+llfTEWLJSQt/uRAo2IYQQg/ZZDuEN9NHh7uCXu38XtbMoPX09dB0LvfjHkWeG/UTY6u5Ap9Zi1sdP8eimTl58Do3OkWfYunttmHRxIf9EPDkmEbvXgS8Q+bCEYLW62rG6O9iQu35aN/yeDjaXNaHVqFm/JCvSQ4l6CwqSiI3RjLgs8tWP67F0e/jCefM4dWEGKhVsPzx69P5EHU+KPLlga3K20BfwUSQJkcPq9fo5VNvFspLUsPx+kYJNCCHEoF3te0k2JPGfK2/B4XXx4O7f0eEee0P8VGty9u+zuLh4E44+J09XvHjSMVZPJymGZNSqmftPXW58Nu09Vjy+4YMIQh3pPyDZkISCEpFZWK/fy7MVL4372gOFbUF8bjiGJY7p9fr5+GAra+anYYqNfEPoaKfVqFlWkkpZhRV/YOiySEu3m5c/rmPN/HQWFCaTYIphfn4S24+0oyhKyMeSZEgkRqMftmA7HjgiM2zDOVTbic8fYHlJeJbgz9x/xYQQQoyLq6+Hw51HWZm+lOKEQm5fcTNun5sHd/+O9p7oaO46oMnVv/F+bdZqzi84i22tu9hvPTTkGKu7Y8YuhxyQZ8pGQaHZNXwQQXdveAq2SDbP3tzwIW83vMfutr3jel6joxmtWkuGMfSBAOK47YfbcPf62bg8J9JDmTZWzUvH6e7jaH33kK8/+XYFapWKz51dMvi1NQvSaevsob4t9Msi+5MiM4Yv2Ox1JMYkDP7si6HKKq3ExmiZm5cYlvNLwSaEEAKAvZYDBJQAq9KXAVBgzuP2FbfQF+jjl7sfptUVnmU4E9HkaCFOZyRBb+aCwnPIMWXxjyPP4Dq2NFJRFDrcnaTM0MCRAQPBIyP1Y7P12kkMceAIQPKxN21TvWS2p6+HN+o3A1A/yt694TQ6m8mOy0SjDl0zW3GyzWXNZKUYmZsbut5/M93i4mT0WvWQZZH7qjrYU2Hl4tMKSTYbBr++al4aGrWK7UfCEz6SFZcxbFJkja2eQpldG1ZAUdhb1cGS4mS0mvCUVlKwCSGEAGBX215SY1MGiwCAvPhs/nPFLQRQ+OXu39HsHDnyeSo1uVrIMWWjUqnQqrVct+AqnH0u/nX0BaB/ttDj753xM2yJMQnE6Yw0DlO8+AN+HF4nCSFsmj0gKSYRYMqj/d+qfw+3z02GMX3UsJVPUhSFRmfzYCsEER71bQ5qWuycuTxH9gmOQ4xOw5LiFHYftRBQFPp8Af7x1lEyk42cvyZvyLHxRj0LCpPYcTg8yyKz4jJweJ1D+jvavQ46PJ0SODKCulYHdpeXZXNCnw45QAo2IYQQOLxOjnZXsTJ96UlvtLJNmXxjxa2oVWp+uSfyRVtACdDibCXHdDwqPC8+h08VnM2Ott3stRzE4u6PV041zOyCTaVSkWfKoWGY4BG714GCEpYlkTqNjni9aUqXRNq9Dt5teJ9V6ctYlbGMth7LiHv3Pqm714arr0f6R4XZlrJmdFo165dIjP94rSpNo9vppabZzmvb62nvcnPtefOGnbE5ZX4GVpuH6hZ7yMeRFdf/d3fiLJs0zB5dWYUVlQqWzAnfig4p2IQQQlD2ieWQn5QZl843Vt4KSn//s0iyuDvwBvrIiRuaQLep8GxyTFk8Uf4MdY4GYOb2YDtRbnw2Lc5W/AH/kK+Ho2n2iZJjkgaTOqfCa7Xv4FP8XFR8PvnxOSgoNDmDayI8EDgiM2zh4/H6+OhgK2vmpxNnkLCR8Vo6JxWNWsUbOxp4eWstq0rTWFQ0/AdOK+elotWo2H4o9MvUs4dJiqy116NRaYasvhDH7a20MjcnIawhO1KwCSGEYHfbXjKMaeSYRo7hTjemsjBlPuVdlQSUkZu8htvAm/Sc+KFj1aq1XL/gc7j6eni+6lVg5vZgO1GeKRuf4qe1Z+ibN5u3v2ALxwwbTG0vtg53Fx80fcy6rNWkG9MG3zjWOxqDen6j49j3jElmfsJl++F2PF4/G5dLUTwRRoOWhYXJ7DjSDiq4+uyRewUaDToWF6Ww40gbgRAvi0yMScCgMQwp2GpsdeSastFrpBD/pE67h/p2J8vmhm85JEjBJoQQs56t10FFdzUr05eNue+kNLkEZ59r2BSxqdLsbEGFiixjxkmP5cZnc0HhOXj9Xsz6ePQafQRGOLVyB4NHhu7pGoi9TwzDHjboDx7p9HSFZR/NJ71S8yYqlYoLCs8F+l+TWR8f9D62RmczabEpGLSGsQ8WE7KlrImc1DhKciRsZKJWlfYnmF60rpCUhNG/V09ZmE6300tFQ3dIx6BSqciKSx/8He8P+KmzN8j+tRHsrexPUF5eEgUF24svvsiFF17I+eefz9///veTHq+urua6667jkksu4aabbsJmm/q+LEIIISZmj2UfCgor05eOeez8pP546fLOinAPa0SNzhYyjGnoRvi0d1PB2RTE582a5W/pxlT0ah2Nn0iKtPXaUavUmHRxYblusiGJvkDfYDJnuLS62tjWuosNOeuGRIrnxeeMq2CbLd8PkVDX6qCmxcHG5dkSNjIJ6xZlctOnF/CpU8cujpaXpKLXqtl+ZOxlkS0drnHNxGWdEO3f7GrFG+iT/msjKKvsID0plsxkY1ivM2bB1tbWxoMPPsg//vEPnnvuOZ566ikqKysHH1cUhf/4j//g5ptv5oUXXmDBggX8/ve/D+ughRBChM7utn1kxWWQHcRysSRDIumxqZR3VY55bLg0O1tGXbqpUWv4xspbuXnJ9VM4qshRq9TkmLJpcA4tXmy9dhL05rA1Dh/sxdYb3qTIF6vfQK/Rsang7CFfz4/PocXVhtfvHfX5bp8Hq7tDAkfCaEtZU3/YyGJZcjoZOq2a05ZkBRUNb9BrWVqSys4j7Sc13D7R8x/U8N0/bONAdUfQ48gyZeLsc+HwOo8HjiRI4Mgn9Xr9HK7rYnlJatg/qBjzO2Lr1q2sXbuWxMREjEYjmzZt4rXXXht8/ODBgxiNRjZs2ADArbfeyrXXXhu+EQshhAiZ7l4b1bbaEcNGhjMvuYSK7uqTQi6mgtvnpsPTNWrBBqDX6GbVfou8+GwaHS1D9hbawtQ0e0DyYLR/d9iuUWdvoMyyn3PyNmDSD50pzAsyeGTgcZlhm5jnP6jhl//aS32bY9jHPV4fHx1q45T56RglbGRKnTI/HUdPH0c+0XB7wPMf1PD8BzUAWLo9QZ83azB4pJUaex3xehPJhqRJj3emOVjbic8fYFmYl0NCEAVbe3s7aWlpg39OT0+nre343oX6+npSU1P53//9Xz7zmc/w/e9/H6MxvNOCQgghQmN3+7HlkBnBF2ylSSX0+r2DSYxTqelYS4GxCrbZJteUjcfvocN9fLar22sPW0IkMPgGLpzNs1+sfp04nZGz8zec9Fh+fC4wdgPtwZAa+Z4Ztz5fgNe317OvqoN7/ryDx149gs01dEZz26E2er1+Nq6QBMGptnROCjF6DdsPnbyn+IVjxdr6xZlo1Cq6ncG1wIDjBVuzq41aWz3F5gJZ6jqMskorsTHaKWkSrx3rgEAgMOQvSVGUIX/2+Xxs376dxx9/nCVLlvDLX/6Sn/zkJ/zkJz8JehApKaZxDluMJi0tPtJDmHHknoaW3M/QGQh8mOg93bf3AIWJuSwuKA76OevNy/jTgb/T0NvAqWlLJnTdidrd3V+QLC2YS4oxfN9H0+17dIlmLpSDXd3JwrRCAOxeO8uzFoTttaQqJvQaHR6Va8xrTGQMh9qPcrjzKNct+yz5WWknPZ6qmDDHmLD0tY96fmuthXh9HHNzc2fUm86p+B7dfqgVj9fPN69dRWVjNy++X82OI+1cde48LjmjGL1Ow4cHWinMMrN22fRulj3dfuYHrFucxc7DbdyRFIdO2z8P88Qb5Tz3QQ1nr87j9s+t4Ms/fAOPLxD0a0xVTMTpYmnoaaDdbeXcuadP6P5M13sajEBA4UBNJ6sXZJCVGQUFW2ZmJjt37hz8s8ViIT09ffDPaWlpFBQUsGRJ/z/aF110Ebfffvu4BtHR4SQQCH/K1GyQlhaPxTL8sgUxMXJPQ0vuZ+i4fR7u3/Erzp17Oqelrh/38zvcXVR01HBJ8afG/XeSa8piT+NBNqafMe7rTkZ5aw1GbSx+pwaLKzzfR9PxezTWH49apeZgUxVzDHPp9Xvp6XMTE4gN62tJikmkqat91GtM5H4qisJfdz9LYkwCKxNXjvj8nLhsjlpqRj1/paWO7LgsrFbnuMYQzabqe/SdbXUYY7TMy45nYV4Cp5Sm8c93KvnLy4d4+YNqNi7PprLRxrXnzZvW93c6/swPWFqczObdjWzZUceyklRe+LCG596v4bTFmXz+7BI6O5zEG3W0Wl3jeo0Zxgx2NO0FIF2TOe77M53vaTCqmm10O3qZn5cQktepVqtGncAac0nk+vXr+eijj+js7MTtdvPGG28M7lcDWLFiBZ2dnRw5cgSAd955h0WLFk164EIIIUZ3qKOcdreVf+x7jp1tZeN+/h7LPgBWjWM55IB5ySXU2OrGDHwItaZjgSPT+ZP8cNBpdGQa0weDR2zHIv3DuYcN+gu2UO9ha3G18dCeR6i21XFh4bmj7kXMj8+lxdVGn79v2Mf9AT/NrlbZvzYBPn+APRVWVsxNHQzByEw2cvsVS/nvq5cTo9fwzJZq9Fo16xad3GJDTI3FRckYY7RsP9w+pFj70oULUKv7f08mmmKwjWNJJPQvi/QrftQqNfnm3HAMfVrbW9mBWqViSXHKlFxvzBm2jIwM7rjjDq6//nr6+vq44oorWLp0KTfffDO33347S5Ys4Te/+Q133nknbrebzMxM7r///qkYuxBCzGr7rYeJ0xnJT8zmb4f/SbIhieJxJHntattLfnwuqbHj/wenNGkub9e/R5WtlgXJ88b9/IkIKAGaXK2sz1ozJdebbvLiczjceRToDxyB8BdsyYYk9nccCsm5ev1eXq15i7cb3sOgieGa0stZn33KqM/Ji88hoARodrVSYM476fG2Hgu+gE8SIifgcF0XPb0+Vs1PP+mxRYXJ3P2lNWw90EqMTiNhIxGk1ahZWZrG1v2tBBTlpGINIMEUQ0Xj+FpuDexjyzFlETML+lmO195KKyW5CZhip+Z7f8yCDeDiiy/m4osvHvK1P/zhD4P/vWzZMp5++unQjkwIIcSI/AE/BzsOsyR1ITevvZrvvP4THtn3GN9a/XVSY5PHfL6lp4N6RyOfKfn0hK5fkliEWqWmvLNyygq2DncXXr9XwiNGkBufzbbWXdh6HXQfK9jCGToC/c2zHV4nff6+EfvijUVRFPZaD/L00Rfo6u1mXdYaLp1zAfH6sfe35x9rGl7vaBq2YGt09vemkxm28dt5pB2DXsOiwuF/n2jUas5YKvc1GqxbmMEH+1qGLdYAEuP0ON19+PyBoFoGwPGCTfqvnazD5qGh3clVZ5VM2TXD05xFCCFEWFXbaunxuVmSuhBzjInbln4JvxLg4X1/xu1zj/n83e39exNWpI3dLHs4MRo9Reb8Ke3H1nTszbcUbMPLO1aUNDqbsHmnZoZtoBdbV2/3hJ5vdXfw8L4/84f9fyVWa+C/Vt7GFxZcGVSxBv0zfEZtLA2OxmEfb3Q2o1VryTCeHFoiRubzB9h91MLyuamDQRYiei0oTOa+m0/lS58+uVgDSIyPAcDmDH4Je358DvE6E4tTF4ZsnDPF3iorAMtKpmY5JEjBJoQQ09I+6yG0Kg0LkucCkBGXzs2Lr6O9x8KjB/4+ao80S08H21p3U2TOJyV24r11SpNKaHA00dPXM+FzjEeTswUVqsFPfsVQA8v+GhzNdPfa0Gv0GDSGsF5zINp/IvvYLD0d3LftF1R2V/PZkov4zpr/ZE5i4bjOoVKpyI/PHTHav8nRQnZcBhq1Ztzjm83KG7pxeXysLj15OaSITlkpcahH2NubENe/pHE80f5GnZGfnPE9FqWUhmR80aitqwdHz/j3Ye880k5GUixZKXFjHxwiUrAJIcQ0oygK+62HmJdUgkF7/A15aXIJV5d+hsOdR3m64oXByH/oX0K5u30fv97zB+7++KdY3FbOyjt9UuMoTZ6LgsLR7upJnSdYTa5W0o2p6GU/xbBitbGkGpJpdDRh67WTqDeHPZwleWCGbQIF237rQbyBPv5n9e2cnb9hwkVVXnwOzc5WfAHfkK8rikKjs1mWQ07AriPtxOg0LC4ae3m1iH6Jpv4Ztu5xzLDNZIqi8ObOBu78wzYeeLIMfyAQ9HP3Vlo5Ut/NmVPcdzCoPWxCCCHCJ6AE2NO+j0UpCzBoY8Y8vq3HgsXdwdl5J0fqn5Z9Km09Ft6uf490YxpLUhfyYfM2PmrZgcPrJCkmkYuKzmdd9hoSYybXO6bQnIderaO8s5LlaYsnda5gNDmayZO0slHlxufQ4GzGrI8P+3JIgISYBFSo6JzAksgjXZVkGNPIjJvcLE5efA5+pT8NcqCZNoDNa8fZ5yJHAkfGJRBQ2HXUwrKSFPQ6mZmcCRJN459hm6ncvT7+/OoRdh5ppyAjnro2B2/vbOT8U8beq9fnC/DE2xVkJhs5Z9XU/lskBZsQQkTYC1Wv8Wb9Zs7J28Dlcy8a8/j91v5UviUj7C24bM6FWHo6eKbiRZ6peBGAxanzOT17LQtTSlGrQrO4QqvWUpJYPCX72Dw+D1ZPJ+uyJSFyNHnx2ZRZ9uPxeZh/bLlsOOnUWsx6E52ernE9zxfwUdFdzdrM1ZMew0CR1uBoGlKwNTokcGQijjZ04+jpk+WQM0i8UY9apcLmmt0FW2O7k988d4D2rh6uPHMOm07N56F/7ePfH9Swen46yebRl5C/tbOB9i43d1y1LOjwllCRJZFCCBFB21t382b9ZmK1Bj5s3o7b5xnzOfuth8gzZQ8GPnySWqXmi4uu4ZTMlVxQeA73rv9/3Lr0SyxOXRCyYm1AaXIJbT3tdPeOLzJ6vJpdbYAEjoxloDhx9rkmPYMarGRD0riXRNbaG/D6vZQmTz5lLTU2mVit4aR9bI0SUjMhO8vb0WvVU9ZfSoSfWq3CHKej2zF7l0RuPdDCD/+6E3evj/+5ZgUXrC1ArVJx7fnzCAQUnni7YtTndzl6eWFrLctLUiPysyEFmxBCREitvZ6/H3mauYnF3LbsJjx+D1ubt4/6HIfXSbWtbsTZtQExGj3XL/wcny4+f8TCLhRKk/rfcJd3hneWbSAhMjtO3nyPJi/++L6KqVgSCf1JkeMt2Mo7K1ChYl5i8aSvr1KpyDPl0HBSwdZCamwKsdrwBq/MJAFFYVe5hSVzUojRy3LImSTRFEP3LJxh6/P5+ctrR/jjS4cpyjJz95fWUJp/PGwrPTGWi9cXsqvcwt5K64jneXpzFX5/gKvPmboo/xNJwSaEEBHQ3Wvj9/v+QoLezJcXX0dxQgFzE4t5t+GDURMeD3YcQUEZs2CbKjmmLOJ0xrAvi2xythKrNQyGXIjhJcSYByPxp2qGLcmQSGdvNwEl+I37R7oqyTfnYtQZQzKGvPgcmpwtQ352mhzN5Mrs2rhUNtqwubyyHHIGSjTFzMoZtt/8+wBbypq5cG0B37xm+WAAy4k+dWo+WSlG/v7mUXr7Tv73t7LJxkcHW9l0Sj7pSaH5nTVeUrAJIcQU8/r7+P2+v+Lx93Lr0i9i0vdHA5+Tv4Gu3m72tO8b8bn7rYdJ0JuHzKREklqlZl7iHMq7KoekUoZak7OF7LissKcezgR5pv7vjQT91MywJRuS8AV8OPtcQR3v8XmotdcPzs6GQn58Dr6Aj9ae9sFrWNwdsn9tnHaWt6PVqFk6R5ZDzjSJJv2s28PW5wtwoLqT89fkccWZc9Cohy97tBo1128qxWrz8NLW2iGPBRSFv795lESTnk+vK5iCUQ9PCjYhhJhCiqLwjyPPUOdo4IaF15Btyhx8bFHKfDKMabzV8N6wxU+fv49DneUsSVsYkcKl1+vnr68d4ZktVXTYju+1K00uobvXhsU98nKS4SiKgtvnpr3HOuqsYkAJ0Oxskb1IQRrox5Y4RUsiU471YmtxtgV1fGV3DQElwPyk0IWiDHyAUW/vb6Dd7GpFQRm8F2Jsg8shi5OJjZFMupkmwRSDo6cPnz/4mfDprqXDRUBRKMoa+3dhaX4S6xdn8tq2epqsxz98+mBfC3WtDq48qwSDPnI/F/ITKYQQU+it+i3saNvNRUWbWJa2aMhjapWas/PO4InyZ6normZe0pwhjx/trsbr97I0Assh7T1eHvrXPmpb7KCCVz6uY+XcNM5elcu81GP72LoqSTemnfTcvoCPD5o+psHRhMPrxNHnxOF14vQ68Sn9hVqhOZ9vrLwVnfrkf5Y6Pd14/L2yvC1I67L6kzTDuXfxRPOSSojR6NnRtieoEJEjXRXo1FqKE0L3aXWaMRWDJoZ6RxPrWCMJkRNQ02yny9HLFRvnjH2wmHYGov3tLu+YaYgzxUDhlZMWXIPrq84qYW+llb+9Xs63P78Cd6+PZ7ZUUZKTwNqFGeEc6pikYBNCiClywHqY56teZUX6Uj5VePawx5ySuYoXq1/n7fr3TirY9lsPodfomZc4tW+o2rvdPPhUGZ2OXr56+RLyM0xs3tPMe3ub2XXUQlaqkdg5Jg5ZKzgjZ93g8xRFocxygOcqX8bq6SQxJgGzPh6zPp4cUxbxOhPxehNefx8v1bzOsxUv8bnSy066fpOzBYBsKdiCkm5M5dI5F0zZ9WI0elakL2VP+z6unHcpMWM0Ni/vrGROQhE6jS5kY1Cr1OTGZw8GjzQ6m4nTGqdsH99MsLO8HY1axbKS1EgPRYRBwgnNs2dNwWZxoVGryEwObt+ZOU7PFWfO4S+vlbP1QCv1bU6cPX3811XzIr4cXwo2IYSYAg6vkz8ffIJcUxbXLbhqxF/+eo2ODTnreKX2LVpdbWTG9X+qpygK+62HWJA8L6RvdMdS1+rgwX/txe8P8M2rlzM3NxGAK86cw6WnF7L9cDtv7WrE2p7A3r5ydh5pZ/X8dOodjTxT8SKV3TVkx2Xy9eU3j9oXzO1z83bDe8xJKGB15oohjzU5m1GhGrJ8VESXtZmr+bhlJ3stBzglc+WIx9l6HTS7Wlnzib/jUMiLz+GDpm34A34aHS3kxGdH/E3WdKEoCjuPWFhUlIzRIG8NZ6KkwYJt9uxja7Q4yUwxjqtn2hnLsvlgfwtPvl2Bx+vnjGXZFGTGh3GUwZE9bEIIMQXKLAfw+D18YcFVY85AbMhdj06t5Z2G9we/1uhsprvXNq50yIO1nVQ22QgEJhYGcqCmg5/8Yzc6jYr/94VVg8XaAJ1Ww2lLsvjeDau5aMkqVNo+Xjywnb8d/if37/g1ra52ri69nO+s+c8xmzhfOucCihMK+Xv5M7S6hu6FanK2khabMuZ9E5EzJ7GQFEMyH7fsHPW4o8fSREMZODIgPz6XvkAfLa42ml0tsoR2HGpbHXTYPZIOOYMlHFsSaZtFBVuTxUVummlcz1GrVFy/aT7uXj96nYbLN06+9UgoyMcoQggxBfZZDpIamxJUcEa83sQpmavY1rqLi4s/RbzexD7rIVSoWJwyP6jrOd19/OLJMhQgzqBlYWEyi4uTWVyUQlL8ybHGn7T1QAt/fuUIWSlx3HHVslGfo1KpOGPOEl5rfQFr8nt0tWo4J38Dnyo8m1htbFDj1ag13LT4Wn68/Zf8Yf/f+Nbqr2PQ9l+z2dkiyyGjnFql5tSsVbxa8xadni6SDUnDHlfeVYlRGxuWlNOBc+5sK6Mv4JP9a+MwsBxy+VxZDjlTmY16VCrocs6OaH93r48Ou4eNqeP/PZCXbuLWSxcRF6vDbIyODwqlYBNCiDBz+zyUd1VyZu5pQS/ROifvDD5s3saWxq1cVHw+B6yHKErIH+yxNZaGdicKcOHaAuwuL/trOthxpD/yPDfNxOLiZBLj9CjAQCClggIKdNp7eXt3I/PzE/na5UuDWiKVGJNAUWwplU3d3LDick6ZUxTUOD95ji8t+jz/V/ZHnix/lhsWXo030IfF3THqMjsRHU7NXMUrNW+yvXU3nyo856THFUXhSGcF85JKUKtCv8Anw5iGXq1jW+suAEmIDJK718cH+1pYVJSMKXbqlluLqaVWqzDH6WfNDNtA4Mh4Z9gGrJ4fXbPNUrAJIUSYHeoox6/4WfqJVMjRZMSlsyR1Ae83fcQpmSuodzSNK0iisd0JwHlr8kiI06MoCo0WFweqO9hf3cGbOxrwj7JUcu2iDL50wQJ02uDfWN+6/Hpu3/I+7dkqmGAuyvzkuXy66DxeqnmDOYlF5JqyUVBkhm0aSI1NZm5iMdtadrGp4OyTPpywuK109XZzftJZYbn+QPBIta0OrUpDpjG63nBFq9e31+Po6ePS08f/IYuYXhJNMXTPkhm2Rkv/v4HBJkRGOynYhBAizPZZD2LSxY07xvycvA380voIjx18EmBccf4NFidmo46EuP7lHCqVirx0E3npJi5YW0Cfz0+fLwCoGHhfrVKB6tif9TrNuMYKYIrVkZMax9GG7nE/90SbCs+mylbL00efZ23WagDpwTZNnJq1mscP/5Maex3FCYVDHis/tn9tfhDR/xOVF59Lta2OLFMmGvX4v4dnG5vLy+vbG1g9Pz2oXlViekuM09PlmCUzbBYXMToNKQkzIxFTQkeEECKMfAEfB6xHWJK6cNzLwEoSi8mPz6XO0UBabAoZ45gxaGh3kps+8lIQnVaD0aDDaNASG9P/P4NeS4xeM6FibcC8vEQqm2z4AxNvzqpWqfniwmuI18fzQfM2DJqYwebMIrqtSFuCXqMfNnzkSGclSTGJpMWGb5/UwD422b8WnJc+rKXPF+DyDdERrCDCK8EUM2tSIpssTnLS4lDPkKRYKdiEECKMKrqq8fg9JzXJDoZKpeKc/A0ALEldGPT+N38gQLPVRd4oBVu4zMtLxOP103BsSeZEmfRx3LT4WjQqDdmmLIlnnyYM2hhWpC1hV9s+vP6+wa8HlABHuyopTS4J699lQXwuIPvXgtHe1cPmsiY2LM8Ouk+VmN4STXocPX34/BP/QG06GNgCkJM6M5ZDghRsQggRVvusB9GrdZQmjR5rP5IVaUu4oPAczsw9LejntHe56fMFJrzZejLm5SUCcLS+e9LnKkoo4D+WfonPzr1o0ucSU2dt1io8fg/7LAcGv9bgaKLH52b+BH8OgpVtyuQrS25gfdaasF5nJvj3+zVoNCouOa0w0kMRUyTRFIMC2F0zex+bvacPp7svIv8GhosUbEIIESYBJcA+6yEWpJSin2Cza41aw0XFm0iJTQ76OQOzW5GYYUuKjyEt0cDRRltIzrcgZR6F5vyQnEtMjZLEYpINSXx8LK0Rju9fKw3j/rUBy9IWoZeefaOqa3Ww7VAb563OI9E0dpsPMTMM/F3bZnjBNtMCR0AKNiGECJsGRxPdvTaWpY5/OeSkrtvuRKNWkZUSmX+s5uUlcrShG0WZWMNuMb2pVWpOzVzJkc4Kunv7C/fyzkqy4zIx6+MjPDoB8PSWKuIMWi44dXxBSGJ6G2ie3T3Dg0eaLJOL9I9GUrAJIUSY7LMcRK1Ssyg1uGbXodLY7iQzxTiuSP5QmpebiNPdR3NHT0SuLyLv1MzVKChsb9mN199Hla1mSmbXxNgO13ZysKaTi9YXBtVjUcwcAzNs3bNghi3eqMMcN3Nm2qVgE0KIMNlrPcichEJMuqmd6WqwOMmL4CeL8/ITASYd7y+mrzRjCnMSivi4dSfl1ir6Ar6w718TY1MUhX9triLZHMPZK3MiPRwxxcxxOlQw45tnN1lcM2p2DaRgE0KIsGjvsdDiamNZ2uJJnUdRFLYdaqPH0zf2wYDL00envTci+9cGpCfGkmDSUyEF26y2NmsVbT0W/n3oNdQqNSWJ0pg5FA7XdlLTPLE9ojvLLdS2OvjMGcXotNKnbrbRqNWY4/QzOto/oCg0W2dWQiRIwSaEEGGxz3oIGF+z6+HsPmrhkRcO8vauxqCObzwWODJaD7ZwU6lUlOYlUi772Ga1FelL0al1HGgvp9Ccj0E7MxrYRpKiKPzuhYP8v99+SJPVNa7n+vwBnt1SRU5aHOsWZYZphCLaJZj0dDtn7pJIq81Db59/RgWOgBRsQggRFvssB8kxZY0r3fGTfP4AT2+pBuBwXVdQzxlIiIz0cpC5uYl0OXqx2jwRHYeInFitgeVpSwCYnyT710LBavPg6OnD5e7jl/8so2sc4RHv72uhrcvNZzfMQa2WvoazVeIMb57dZImOfwNDTQo2IYQIMYfXSbWtbtLpkO/va6Gts4f8dBOVTTZ6+/xjPqfR4sQUqyPRFNnN1qUD/dhkWeSsdnrOqWhUapZMcqZZ9KtpsQPwtSuX4/T4+OW/9uLu9Y35vK0HWnjiraPMy0tkWUlKuIcpoliiSY9tBs+wNR5LiMyWJZFCCCFGs996CAWFpZPYv+bx+nj+gxrm5ibw2TPn4PMrVAbR26yh3UVeugmVKrKfoGenxRFn0ErBNsuVJBbx58/8nHxzbqSHMiPUtNjRatScsyaPr162mCaLi9/8ez++/8/efUbHcZ4H3//PzDYs2qL33tm7KFJUobolWc29yHGLHcdx4rxPqlOc+MnjdCd2Eju2E3e5ybZ6oSiJktjFDpDovfe6fcr7YQmQIFEWwAK7AO/fOTk5XszM3liBwFxzX0XTZzxeNwx+eaiB7z5fTXFWPJ9/bGPYfzcI4eWIsTLm9KHpM//MrHad/RMkx9uIsq6tDqgiYBMEQQixCwMXSbQlkB2TsehrHDjZzpjTx3tvL6Y024EiS1xqHZrzHF036OyfiIhUEFmSKMl2iIBNwGYWtWuh0tw1Rl5aDCZFZkNhEh+7v4xLLcN8/6Wa6+pF3V6V//hVJS8db+P2LZn84fu3EBNlDtPKhUgRH2PFAMacwTWyWm06+9dewxEQAZsgCEJIeVQv1UP1bEpet+gn2WNOHy+dbGNbaQrF2fFYLQqFmXFUt8xdx9Y34san6mSnRsYfq9IcB73D7jXfQloQVoKm67T0jlOQETf12r5NmTxySwFHq3r4zdvNU6/3j7j5fz8+zYXGQT58dykfvbcMkyJu+QSm0uXXYh2bqun0DLnC2nRruayt/UJBEIQwqxmqQ9VVNqcsvn7tuSMt+P06j99WOPVaRV4Czx1pwenxE22b+Sn5ZIfI3NTYRb93KJVO1rF1jLKzPDW8ixGEVa57wIXPr08L2AAe2pvP0LiH54+2kBhnJTMpmv/4dSW6bvDF929mff7iGx8Ja8/k8Oy1WMfWM+RC0w2xwyYIgiDM7cLAJeymKIriFzdzqnfYxaFzndy6OYOMpCt/dNblJ2IAtW0js57b1jeBJEFmsn1R7x1quWkxWM0KdXOsWRCE4DRdbjhSkDk9YJMkiY/eW8amoiR+9Eot//TTs8REmfmLj+0QwZpwncmAbS3usHWs0Q6RIAI2QRCEkNF0jcqBS2xIrkCRFzeU9tdvNqEoEu++ZXrAV5gZh8Usz5kW2dE3QXqiPWIG4poUmaKsOGpFHZsgLFlL9xh2q4nUhKjrvqbIMp99eD2l2Q42FibxF09sJz0xMh7cCJElLtqMxNoM2Dr7nSiyRHrS2vvZFymRgiAIIdI02opLdbNpke38m7vHeKemj4f25E89BZ1kUmRKcxxzNh7p6J+g8Jqn7+FWmuPgmbeb50zlFARhfk3dY+RnxCLPUhtrs5j4kw9vW+FVCauNIsvERq/N4dmd/U7SEu1rsl5z7X1HgiAIYVI7XI+ERNkihgQbhsEv32gg1m7mvptyZzxmXV4i3YOuGYflujwqA6MeciKs2Losx4EB1AcxkkAI3vC4lyOV3eFehrBCfH6Njj7ndfVrgrAYjmjLmmwG1dE/QXbK2qtfAxGwCYIghEzNUD35cTnYzdenLM2nsmmQmrYR3r23YNb5MRV5CYH3ab0+LTJSc/cLMuIwKZJo7x9izx9r4X9eqGZozBPupQgroK1vAt0wKBQBmxAC8TFWRpyrZ4dteNzLgZNtU421ZuL2Bh5arsWGIyBSIgVBEELCrbppGWvn3vz9Cz5X1w2eOtRIqiOK27ZkznpcTloM0TYTl1qHuHlD+rSvTQZskbbDZjEr5GfEiYAthHTD4Fz9AAAtPeMkxok5Z2tdc1eg4Ui+CNiEEHDEWGjrGw/3Muak6wZVzUO8ea6T8w2D6IZBcryNv/nErhkfanYNOoHIe2gZKmKHTRAEIQTqhhsxMChPKFnwuY1do3T0O3lob/6cufeyJFGel0B16/B1Q3Lb+yaItplIiLXOcnb4lOU4aO0Zx+vTwr2UNaG1Z3wqLbalZyzMqxFWQnPPGAmx1oj89y2sPvExVsacPnTdmP/gFTY05uHZw838ybeO8m+/PE9j5yj37srhsw+vZ3DMw09fq5/xvM7+QMCWtUZTIsUOmyAIQgjUDDVgUSwUxM9cfzaXyqYhZElia0nyvMeuy0vgdG0/fSNu0hKudMLq6JsgOyVm0cO6l1NpjoMXjrXS2DXKOtFmfMnO1vcjSxKJcVZaukP7lFzV9DVZsL/aNXeNifo1IWQSYiwYBoy5fNc1uAoXwzD43ks1HKnsxjBgfX4C799fwpaS5KnfSe19E7xwrJWtxclsLU2Zdn5H/wQWs0yyY+ElCauB+K0sCIIQAjXDdZQ4CjHJC38OVtU0SGFWHPYguihWXA54rm7vr+sGHf3OiEuHnFScFY8kwetnOmnoHEXV9HAvaVU7WzdAaU485XkJtPSMX7fbulgXm4f4vX97m87+2etEhJXn9PjpHXZTkBEb7qUIa0R8BA7PbukZ5/CFbvZsSOfvP3sz/98HtrKjPHXaA6SHbykgNy2G779cw+g1NXid/U6ykqNn7aK62omATRAEYYmGPMP0uQYoX0R3yDGXj9aecTYWBLfzlJYQRUKslUtXNR7pGXLi9WtkR2jAFmU1sbM8lTN1/fy/H53m9/7tbf715+d44VgLjV2jaLoI4ILVO+Sic8DJ1tIUCtJjmXD7GQxR45H6jhG8fo1fv9UUkusJoTG5iyp22IRQmdxVG46gTpHHL/ZiUiQ+eGcJqbPskpkUmU8/tB63V+MHL9VMe1jV2T9B1hqtXwOREikIgrBkNUMNAJQnli743EvNQxjAhsKkoI6XJImKvAQuNAaKsGVJouVyQ4JI3WED+OzDG/jQ3T7q2kaoaRumtm2EX70ZCAysFoX37y/m9i1ZYV5l5Dt7udnI1pJkxpx+IFDTlhy/9DSgrkHX1Hs0do5SlBW/5GsKS9fUfbnhSLoI2ITQcMRYACKmtb+m65yo7mVzUfK8mSZZydG85/YifvZaPW9f6ObWzZmMOX2Mufxkr9EOkSB22ARBEJasdrieOEssGdFpCz63smmImCgzeenBpztV5CUw4fZPtThu6R5DkiAzwv9Yxdkt7ChP5SP3lPGVT93Ev/3eLfzOIxtIc0TxwtGWkKX2rWVn6vvJTYshOT6KnNRoFFmipSc0dWzdg07Kcx3E2c386s1G8d8jQjR3jZGeaMduE8/YhdCIiw4EbJEyPLu6dZgxp4/d64P7G3rXjmwq8hL46cF6+oZdU2ncWRH80HKpRMAmCIKwBLqhUzNUT1lCyYIbfuiGwcXmQTYUJC4o736yccely3VszV2jpCXYsZqVBb1/uMVFW9hZnsod27IYHPNOdfm6EQyOejhS2c3/PH+J/3nhEnoQwdGo00djxyjbSgLF9maTQlZKdEgCNk3X6R1ykZ8Rx4N78qlpG5n6+RLCxzAMmrtFwxEhtEyKTKzdHDE7bMeqerFbTWwqCi7TRJYkPvGuCmRZ4rsvVNN++eHlWt5hE49rBEEQlqBroocJv5PyxIXXr7X3TjDm8rOhcGGdExNiraQn2qluHea+m3Jp6R6L2Pq1YGwqSgZqOd84sKq/j7mMTHipaR2mpm2YmtYR+kbcQCAd1OvT2FiYxK6KuZ8un28YwAC2XNVNND89ltO1/RiGsaQOoQMjHlTNICPRzu716bxysp2n3mxkXX5CRHYevVEMj3sZdfpEwxEh5Bwx1ojYYfP6NM7U93NTRSpmU/APHZPibXzk7lK+8/wlugecxESZp3YO1yKxwyYIgrAENcOBmTDliQufv1bZNAjA+oLgniperSI/gbr2ESbcfnoGXRFdvzafhFgruWkxnG8cDPdS5nTwVDvffLpqwamCB95p5w//4wjffu4S79T0k5kczQfuLOFvPrGL//iDfWQlR/PM4eZ5ZyKdresnOd427b91fnocTo/KwOjSGo90X65fy0iOxmySeWRfAa0945yu7V/SdYWlab5cv1aQKXbYhNCKj7EwEgE7bGcb+vH6NG5en77gc3evT2NHeSpOj0p2SvSafrgkAjZBEIQlqBmqJ92eisO68AYNVU2D5KbFEL+Ip4IVuQl4/RqHL3QDkLPKu2NtLkqmsXOUCbc/3EuZ1dsXunmnpo/GzuCHVftVjeePtlCa4+CvfmsH3/j9fXzhPZu4Z2cOOakxKLLMw7cU0D3o4sSl3lmv4/GpXGwZZmtJyrSbksnax9YlpkV2DwbSUTOTArP9bl6fTmZyNL9+q0l08Qyj5u5xFFkidxU/kBEiU2CHLfwB2/GLvSTGWSnJcSz4XEmSeOLeMpLirBRnL/z81UQEbIIgCIvk1/w0jDQvanfN5VFp7BpjY5DdIa9VnpeABBw83Q5Adurqzt3fXJyMYVzZdYw0Vzd5efVUe9DnnbjUx4Tbz8N788lPj0OWr38CvK0shZzUGJ450jxrcFTVNISq6WwrnT5cPTslBkWWaO4JPoicSdegk/hoy1SHNlmWeOzWQnqGXByt7FnStYXFa76c7ryQVDFBCIYjxsKY0z/vzv5yGnP5qGoa4qZ1aYuenxYTZearn7mZR/cVhHh1kUUEbIIgCIvUPNaKX/cvKmCrbh1G0w02BDl/7VoxUWZy02IZGvMSbTORFGdb1HUiRX5GLHF2M+cbBsK9lBnVtY9gEBgCfrq2n6EgZp8ZhsHB0+1kJUdTnpcw63GyJPHILQX0Dbs5WjVzcHS2vp+YKDPF2dN3cs0mmeyUmBDssLnIuLy7NmlrSTKFmXE8c6QZv6ot6frCwumGQUvPGIWi4YiwDBwxVnTDYDyMWQ3vVPehGwY3r1t4OuTVTIq8ptMhQQRsgiAIi1Y9VI8syZQ4Chd8blXzIDaLsqRZVxX5gSAgPzN+1f+xkiWJjUVJVDUNRWQKXk3bMBaTzCceqMDA4I2znfOeU98xSlvvBHfuyJ73v8+WkmTy0mN57kgLqjb9+1c1nfMNg2wuTkKRr/+znZ8RS2vP+KLb8BuGQfegk4xrOqxJksTjtxYyNObljTPzf79CaPUOuXB7NdEhUlgW8dGB4dkj4+FLizx+sYfslJg122wqlETAJgiCsEi1Qw3kx+ViMy1sd8swDKqaBqnIS8CkLP7X8LrLuzb5a+SGbnNRMi6vSkPHaLiXcp26thGKsuJJT7SztSSFN8914fPPvet08HQH0TZTUE+PJUni0X0FDIx6OFzZPf2920dweVW2Xm7nf6389FicHpX+RTYeGZnw4fZqZCZdn1ZbkZ/I+vwEnj/WiturLur6wuI0dV1uOCI6RArLwBF7eXi2MzwBW9+wi8auMW4OcvbajU4EbIIgCIvg9LtoG+9YVDpk96CLwTHvouvXJpXkOMhNi2HnurXxB299QSKKLEVct8gJt5/2vgnKch0A3LU9mwm3n+NzNAkZGvNwpraffZszsVqCqz/aWJhEUWYczx9twa9e2WU7Wz+AxSSzfpb02fz0QMDe0r24OrbJhiPXpkROeuy2Iibcfg68E3ztnrB0zd1jWC0KGTME0oKwVI7JHbYwtfY/frEXCbhpjfz9Wm4iYBMEQViEuuFGDAzKExYesFU1DwEsun5tktWs8OWP72J7+dr4gxdlNVGa4+BChAVs9Zfr18pzAzuaZbkOslNiOHiqfdY0xDfOdmJgsH9rVtDvI0kSj+wLpCC+db4LCOzGnq3vZ31B4qyD0bNSojEp0qIHaE+19J8lMCjIiGN7WQovn2zjXMPAolMvhYVp7h6nID12xkY1grBU8TGBHbZwdIo0DINjl3opy3WQuMrrr1eKCNgEQRAWoWaoDptiJT8uZ8HnVjUNkpFkJ9kRtQwrW902FyfTNeCk//Jg6UhQ2z6C2SRP1RJJksRdO7Lp6HdS2zZy3fE+v8ab57rYWpKy4P/G6/ITKMmO54VjLfj8Gm29EwyNeWdNh4RAwf1SGo90DzqJsio4YmYfL/HeO4pxxFj5+lMX+Kefnp2aDyYEOr6GutOeX9Vp7xtfM+nOQuQxKTIxUeaw7LC19IzTO+Ri9yJmr92oRMAmCIKwCDXDDZQkFKLIC2u37fNr1LaPsGERw7JvBJuLAp9LJO2y1bQNU5QZh9l05U/m7nVpxESZZ2zxf/xSLxNuP3fvyF7wewVq2QoZmfBx6FwXZ+r6kSTYXDz3z0t+Rhwti2w80j3oIj1x7qGzqY4ovvLJXXzknlI6B5x85Qen+NYzVREVWIdD54CT//NfR3j+WEtIr9vRP4GqGaJDpLCsHDEWRsOww3bsYg8mRWJH2ewPooTpRMAmCIKwQAPuIQbcg5QnlC743Nr2EfyqzobCpaVDrlVpiXbSEu0R097f6fHT3jsxlQ45yWJWuG1LJucaBqYFLYZhcPBUB9kpMZQuYhAsBGbslec6ePFYC+/U9FGa7SDWPvdw9fz0WNxelb5FBFBdg86pgdlzMSky+7dl8/efuZkH9+Rzrn6AP//2cX72Wn1EDzxfLh6fyn/9phKPT5tz6PliTO5gig6RwnIKx/BsTdc5eamXzUXJU3MfhfmJgE0QBGGBaofqAShPLF7wuZVNg5hNMmWLvJm/EWwuSqKmbRiPL/xdCSfnr002HLnaHVuzkJB4/UzHtOM7+ie4K4hW/nN5ZF8hYy4/PUMutpbO/xQ6Pz3QSbCle2FpkS6PyuiE77qW/nOJspp47NZCvvqZm9mzIZ1XT7Xz598+zqgzPM0LwsEwDL7/Ug09Qy52VaTSPeiia8AZsus3d40RZzeTGGcN2TUF4VrxMZYVT4msbhlmzOUX6ZALJAI2QRCEBaoersdhjSfNnrrgc6uahijLcWCZpYGEEAjYVM2gumU43Euhtm0EkyJTmHn9TkdinI0d5Sm8db57Krg8eCrQyn/3EjufleY4prpCbi1Jnvf4zORoTIq84Dq2+TpEziUh1srH31XB7z22iQm3n9aeG6eu7fUznZys7uPRfYW8f3+g8dCZuv6QXb+pe4yCjLhVP19RiGyOGCtjTh/6CjYSOnaxB7vVxKYiURawECJgEwRBWADd0KkbaqAsoXjBN1MDI256hlxsWGI7/7WuJMdBlFXhfGP40yJr20YozorDbJo5wL5rew5ur8qxqh4GRt2cqe/nti1ZIQnIP3ZfGZ9+aB0pQTQuMSkyOakxtCwwaOq6HLDNNIMtWLlpgaG3Q2EcwLuSGrtG+dlr9WwuSuJdN+eREGulKDOO0yEK2MacProHXRRnx4fkeoIwG0eMFU03mHCtTEpz37CL03X97ChPmVYTLMxPfFqCIAgL0DHRhVN1LWr+2mQ7/42ifm1OJkVmfUES5xsHw9pC3uXx09Y7Ttk19WtXK8qKIz89loOnO3j9dCcSEncsoJX/XJLjo7h5AWlD+RmxtPaOL+hpefegC5MikexYfGvt+BgLkgTDY2s/YBt3+fjm01UkxFr51EPrkC8/tNlWlkJrzzgDo0tvwlLTFthZLs+b/edOEELBsYKt/b1+jf/4dRVmRebBm/OX/f3WGhGwCYIgLEDDSDMApQlFCz63smmQpDgr6YkLTz+70WwuSmJ0wkdb70TY1lDXMRqoX5uj3lCSJO7ekUP3oItXT7WzrTSZpPjwzBXKT4vF7dXoGw4+aOgecJKWaEeRF387oMgyjhgrw2t8h03XDb7z3CXGnD4+9+gGoq9qmLDtcp3hmbql7wrXtA5jsyhTdYmCsFziYwI1koOjnmV9H8Mw+MHLNXT2T/CZd68XI20WQQRsgiAIC9A40kySLRGHdWHpSqqmU906zIbCJFGXEoSNhUlIENZukbVtw5gUmaKsuTv17ShPJS7agqYb3LVj4XP5QmVyZlfLAmakdQ+5Zh2YvRAJsVaGx5f3pi/cnj/aQlXzEB+6u5T89Ok/E2kJdrJTojlT27fk96luG6E0x7GkIFoQgpGZZCfaZuJ/XqgOaQ3mtV473cHxi708sq9AlAQskvhtIAiCECTDMGgYaabYUbDgc+vbR/D4NDF/LUhx0RYKMuM4H8Z5bDVtI5fnr81dj2Y2yTx2ayG716VREsa6o8xkO2aTTEuQjUf8qkb/iDuolv7zSYi1rukatqrmQZ453MyeDenctjlzxmO2laZQ3zG6pG6Zw+Neeodc142REITlYLeZ+avf2klaYhT/8etKnjxYh6rpIX2PuvYRfv56A1uKk3lgT35Ir30jEQGbIAhCkHpd/Uz4nYsK2M7WD2A2yWwoEPVrwdpclERz91hY2sW7POrl+jVHUMffujmT3373+rDuniqyTG5qTNABW++QG8OA9FAFbGPesNYcLpfTtX186+mLZKZE89F7y2b9b7y9LBUDOFe/+J2KmtZA/VqFqF8TVkiKI4o/+8h27t6Rw8FTHfy/H51e1DzHmYxMePnm01Ukxdv41IMVUzWfwsKJgE0QBCFIjZfr14oWGLAZhsHZ+n7W5SVgtYh2/sHaXBxoZ38hDN0i6ztGMAzmbDgSifLSg288EooOkZMSY214/Rpur7bka0UKt1flf164xH/+pooURxRfeHwT1jm6f2anRJPqiFpSt8jqtmGibSZyLnfeFISVYFJkPnhXCZ9/bCN9w27+5nvvcHqJ6b2qpvPNp6tw+1Q+/+hGMSR7iYIK2J577jne9a53cc899/CTn/xk1uMOHTrE/v37Q7Y4QRCESNIw2kysOYbUqPnnYl2tvW+CwTFvUAOQhStyUmNIiLVyrn7lA7bA/DWJohnmr0Wy/PQ4vD6N3iHXvMd2D7qQICRNcCYHPK+VOrbatmH+6n9OcrSqhwf35POlJ7bPO15BkiS2laVQ3TKMy7O4Nuk1rcOU5jjEToQQFttKU/jyx3eSnmjnP39TxU8OLD5F8n+fu0h9xyi/dX852aniAcRSzRuw9fb28rWvfY0nn3ySp59+mp///Oc0NDRcd9zAwAD/8A//sCyLFARBiAQNI80UOQoWnPZ2pq4fiSs7RkJwJEliZ3kqFxoHGXetbFpkTdswhZnxq27AeX5GoLNgMGmR3YNOkuJtIfkeE2InA7bVXcfmV3V++UYD//jkWRRZ4s8+sp3Hbi3EpASXkLS9NAVNNxZVezkw4mZg1CPSIYWwSnZE8Wcf2cY9O3N47UwHhy90L/gaxy/28NzbTdy1I5vd64IfTSLMbt7fQEePHmX37t04HA7sdjv33nsvL7/88nXH/cVf/AWf//znl2WRgiAI4TbsGWHIM7yo+rVz9QMUZccTH21ZhpWtbbdsykDTDY5f7F2x93R5/LT2js/Zzj9SZSTZsZhkWrrnD9i6BlxkJi89HRKuBGyrufFIR98EX/nBKV460catWzL58id2Upy1sCYyBZlxOGIsnKldeFpktZi/JkQIkyLz/v3FJMRaqb5cVxkst1flBy/Xsr4wiffdUbxMK7zxmOY7oK+vj5SUK2k8qampXLhwYdoxP/zhD1m3bh2bN29e1CKSksRWaSilpIjZLaEmPtPQWo2fZ21rNQA7CzaQkhD8+vuGXLT1TfDxB9ct6/e9Gj/TYKSkxFKc4+B4dS8fete6oM97/VQ7qqZzz015C37PU9W9GAbctClzVX6uRdkOOgedc65d0w16h13sWJcWku/RkRCNJIFXM2a8XqR/jsPjHv7fj09js5j4y0/exK4l7Ars3ZzFqyfbiI2PwmaZ9zZrSkvvBPExFrZUpAe1ix/pn+lqIz7P620uSeFcfT/JyTFBZ5acqu7F69f4wN2lZKSHr2vuWjPvbxJd16f9RzIMY9r/rqur48CBA3z/+9+np6dnUYsYHJxA19deZ6lwSEmJpb8/uA5hQnDEZxpaq/XzPNNejU2xYffHLWj9B0+1A1CaubDzFmK1fqbB2l2Ryo8P1HGqsou8IIYJOz1+/utX5wFYlxM/Z6OImVQ2DGBSJJKizavyc81MsnP4Qje9vWPI8sw3WX0jbvyqjsMeuu8xzm6ho2fsuuuthp/PZw834/Fp/OXHdpCRFL2k9a7LieeFIxqHTraxvSy4ulXDMDhX109ptoOBgfmHxa+Gz3Q1EZ/nzHJTozl0poOqur6ga11PXOjCpEiU5yeKz3QBZFmacwNr3pTI9PR0+vuvbO339/eTmpo69b9ffvll+vv7efzxx/nt3/5t+vr6+NCHPrTEZQuCIESWxpFmCuPzkKWFNdc9Vz9ARpKdtBA0drhR3bQuDZMiB11LcehsJ16fhtenLWoYbGXjAAUZcQsO9CJFRV4CXr/GierZ00i7B0LXIXJSYHj26kuJVDWdN852srEwKSRDxEtzHUTbTJypC77LXt+wm+Fxr0iHFCLKZFp4bVvwaZHVrYH634XsLgvzm/fOY8+ePRw7doyhoSHcbjcHDhzg1ltvnfr6F77wBV555RWeeeYZvv3tb5OamsqTTz65rIsWBEGY5NP8VA5cWtb3mPA76Xb2Lrid/4TbT23bCFtLRHfIpYi2mdlWmszxSz341bk7lvlVjVdPdbC+IJHkeBtHqxaW+eH2qjR2jq66dv5X21KSTE5qDM+83Txrh7fuwUAXyYzk0D1ISIyzrcqA7VRtH6NOH3duzw7J9RRZZktJMucaBoPusDdVvxbk3D9BWAnpiXbioi3UtY8EdbzT46etd1z8HC+DeQO2tLQ0vvjFL/LEE0/wyCOP8OCDD7Jp0yY+/elPU1lZuRJrFARBmNXhruN868L3aRlrW7b3aBxpAVhww5HKxkF0w2BrqegOuVS3bMzA6VE51zB3i/8jVT2MOX28a3ceN69P51LL0IKCiPqOUXTdWNU3HLIk8ei+QvpG3BypnHlXsmvQSVy0hegQzkZKiLWuyqYjr53qIC0hig2FoRtqv700FbdXpSbInYma1mEcMZaQjFgQhFCRJInSHAe17SMYQcx2rGsbwUAMfl8OQeX2PPTQQzz//PO88sorfPrTnwbgO9/5Dhs3bpx2XHZ2Nq+//nroVykIgjCL6sE6AC4O1i7bezSONGOSTeTF5SzovDP1/cTHWCjIWF2zvCLRuvxEEmKtc6ZF6rrByyfayE+PpTzXwZ6N6RhGoMV0sGrahgPz1xbYHTDSbC5OojAzjmePtOBXrx9m3T3oJCPEwUFirBW3V8XtVUN63eXU1DVGY9cYd27PDunss/UFCVjNSlDdIg3DoKZthPK8hAWPDBGE5VaW42BozMvg6PwzFqvbhjGbZAozV/fvz0i0sGIMQRCECOLXVepHmgC4tIwBW8NoM3mxOZjl4HPy/apGVdMQW4uTxRDcEJBlib0b06lqHpx1x+xMXT99w27etTsPSZJIS7BTnBXPkaqeoJ4Oe30aRyu72VSSsmrr1yZJksRjtxYyPO7l0NmuaV8zDIPuARcZIWrpP2k1zmJ77XQ7NovC3o0ZIb2u2aSwqSiJM/UD8zZV6xp0Meb0Ub6K03CFtWuqji2ItMia1hGKs+Ixm0R4EWriExUEYdVqHm3Br/vJjc2idaydCb8z5O/hUb20j3cuOB3yUsswXr/GFlG/FjJ7N2ZgGHC06vpdNsMweOlEK6kJUWwrvfKZ79mQTteAk9be+buVvXG2kzGXn/fdWRrSdYfLuvxEynMdvHCsBa/vyi7bmNOHy6uSkRTaHbbVFrCNTng5Wd3H3o0ZRFlD3yBhe1kKY04fDZ2jcx5Xc3nOlUgjEyJRZko00TbTvAHbuMtHR/+EaJyzTETAJgjCqlU9VI8syby76H4MDGoup0eGUstYG7qhL7jhyNn6AawWRdyEhVBagp3S7HgOV16/Y1bbNkJz9zj37cqd1sp+Z0UqJkXmaOXcaZFen8ZLJ1pZl5/A+sKkZVl/ODx2WxFjLj8HT7dPvdZ1ueFIKDtEQqDpCMDQ+PypU5HgzXNdaLoRsmYj19pYmIRJkXn5RBv6HDu8Na3DJMXZSHFELcs6BGEp5Mt1bHVtI3MeV3v56xVip3hZiIBNEIRVq2aonvy4XMoSiok227k0FHzANuFz8puGF3Crc99cNow0IyFRGB/8AGbdMDjXMMDGwiSRGhJiezdl0Dvkum7X4sUTrcTZzezZMH3gcbTNzJaSZI5f6p2zY98bZzsZd/l5+JaFBeaRrjgrnk1FSbx0vA2Xxw9Az2BgJzrUO2yOmNWzw3Z1K//lavQRZTXx3tuLONcwwAtHW2Y8RjcMatqGKc9zLMsaBCEUSnMc9I245/y3Xd02jNWskJ8hBpAvB3EnIQjCqjThd9I+3klFYgmyJFORWMqlwVp0I7g22m91HuVg25s83fjinMc1jjSTHZtJlMkW9NqausYYc/rYViK6Q4bazvJUrGZlWvOR9r4JqpqGuHNHDpYZas/2bEhnwu2nsmlwxmtO7q6tz0+gJNuxXEsPm8duLcTlVXn5ZGCXrWvQhc2iTKUwhorZJBNnN6+KgC3Urfxnc9eObG5en8bTbzdzfoYOpx19Ezg9qtiJFyJa2eWuubXts3c9rWkdpiQ7HpMiQovlID5VQRBWpdqhBgwMyhMD9UbrEssY90/QMdE1z5mBeqcTPWeQJZnDncdpGGme8ThVV2kea6M4fqHpkP0ossSmorWTWhcpbBYTO8pTOFnTN1WX9dKJVqxmhf3bsmY8Z0NBInF286wz2a7srhUu27rDKTctlh3lqbx6qp0xly/QITLJviwdCRNibQyNRX7Athyt/GciSRJP3FdOTloM337uEr1Drmlfn6xfEw1HhEiWkxqDzaJQ1z5zPebohJfuQZeoX1tGImATBGFVqhmqJ8pkIy828IS8IikQuF0Koo6tabSVAfcg7y15N4m2BJ6s+RV+/fpW5O3jnfh1/8Lr1+oGKM1xYA/hjCvhin2bMvH6NE7V9jEw6ubkpT5u25I560wxkyJz07p0zjcMMOH2T/va1O5aQSLF2Wu3FfWj+wrw+TVePNZK96CLjBDXr01KiLUyHOE1bMvVyn82VrPC5x/biCJLfOPXldPGHtS0jZCaEDVV/ycIkUiRZUqyHdTOMlewZrJ+TQRsy0YEbIIgrDqGYVAzXE+powhFDqTAxVliyYnN4tJgzbznn+g5jUU2syt9Ox8oe4xeVx8HWq6fITm581bkyA96bd2DTnqGXNM6FQqhVZIdT2pCFEcquzlwsh1Jgnt2zj0jb+/GdFTN4J3q3mmvv362I7C7tndt1a5dKyMpmj3r03n9TCfD496Q169NSoizRnxK5HK18p9LcnwUv/PwenoGXfzvC9UYhoGm69S2D4ubXGFVKM2Jp/vyCIpr1bQNE2VVyE2LCcPKbgwiYBMEYdXpcw8w5BmmPLFk2uvrE8toHmvD5XfPeq5P83Om7zxbUjdiM1lZn1TGjrQtvNL6Bt3O6TfzDSPNpNlTiLMEX0R9tj5Qp7KlWNSvLRdJkti7MYOathHePN/FTevS5t2hyEmNITslmiNXpUV6fRovHW9b87trk959S8FUd81Qd4iclBhrxelRp40RWCnjLh8nLvXyk1freP1MB73Druu6iS53K/+5VOQn8r47ijhd188Lx1pp653A7dVEOqSwKpRd/jmtm6G9f03rMKXZDhRZhBXLZWV/WwmCIIRAzVA9wFT92qSKpDJebn2dmuF6tqVumvHcyoFLuFUPN6Vvn3rtPSXvpnqwjidrnuKL234HWZLRDZ2m0Ra2pGxc0NrO1veTlxZLUrxIcVpOezek8/RbTfhVnftuyp33eEmS2LMhg1+80XC5hiua1890MOFee50hZ5PiiOLWzZm8cbaTzBAPzZ6UGBv4uR+e8C5b98VJqqbT2DlKVfMQVc1DtPWMYxBIgZ3sCJocb2NDQSLrCxKpyEtY9lb+87l7Zw4tveP85q0m6jpGACi/3NBBECJZfnosFpNMXfsIO8pTp14fHvfSO+zm9q0z1xALoSECNkEQVp2aoXqSbAmkRE1v6lEQl0uUyUb1YO2sAduJntM4rPGUJhRNvRZrieHxkof4YfXPOdx5nFuz99Dt7MWluhc0MHvU6aOpc+yGCQDCKTHOxk3r05CA7JTg0nB2r0/jl4caOHaxh3ftzuOlE21sKEikOGvt765Nes/tRawvSCRtmYKpyc6TQ2OeZQvYfH6N771Uw7n6Abx+DVmSKMqK4+F9BawvSKQgPY7+ETdVzUNcahni+KVeDp3rQpICtTjL2cp/PpIk8bH7yunqd1LVNERGkp34mNB26xSE5WBSZIqy4q8boC0a56wMEbAJgrCqaLpG3XAj29M2XdflTpEVyhNKuDRUh2EY13191DtO9VAdd+XehixNT93Ylb6Nkz1neKbxJTYmr7uqfi344KuufQQD2LCGBi9Hst9+aP2CjnfEWFlfkMixqh6sZuWG2l2bFGU1LWt9ZULc8s9ie6emjxOXerllUwZbi5Mpz0u4Lr0xLdFOWqKdO7dno2o6TV1jXGweoqFzNOz/zSebkHzlh6fYLFKnhVWkLMfBM4ebcXr8U02eqluHibaZyBH1a8tKBGyCIKwqrePteDTPdemQk9YllXO2v5IuZw9ZMdObCpzqPYtu6NyUvu268yRJ4oPlj/F/T/wrv6h7BrNswmGNJ8kW/FPDxs5RzCZZFF5HsD0b0vn2s5d45nAzGwoTKbqBdtdWQsLl3aKhZQzYjlR2k5oQxcfvLw9qNIFJkSnNcVCa41i2NS1UsiOKf/jszVhM188NFIRIVZrjwADq20fZcnnOaE3bMGW5CSvScfVGJqoDBUFYVaoH65CQKEsonvHr66ba+9de97UTPafJi80hPTptxnOTo5J4oOBuLgxc5Hx/FcWOggXNqmrsGiUvPVYMDo1g20pSiLIqqJqx5jtDhoPFrBATtXzDswdG3dS0jbBnQ/qyzJFbSTaLCVle3d+DcGMpzIzDpEhTjUcGRtwMjHpEHeYKEHcVgiCsKjXD9eTGZhNtnrkGxWGNJysm47qArWO8i86Jbm7K2D7jeZP25+wjJyYT1dAoWsDAbL+q09ozTlFmXNDnCCvPYla4/6Y8bt+aJXbXlklirJXhseWZxXbscpfPPRvSl+X6giDMzmJWKMiIo7Y9ULdWfXkumxiYvfxEwCYIwqrhVt20jLVf187/WusSy2gcbcGjXrlpPNFzGkVS2J62ec5zFVnhwxXvIzc2i43JFUGvra13HFUzKMoUQUCke3BPPk/cWxbuZaxZgeHZod9hMwyDI1U9lOc6SI6PCvn1BUGYX1mug9aeCdxelZrWEWLtZrKWqeuscIUI2ARBWDXqhpvQDX3+gC2pDM3QqB1uBAKNSt7pPcuG5ApizPP/YcmJzeRPdv4+CTZH0Gtr7BoDELs2wg0vIc62LDVsDZ2j9A27V3TgtSAI05XmONANg8bO0an6tdWenrwaiIBNEIRVo2aoDotspiA+b87jCuPzsCoWLg0F0iKrh+oY903M2GwkVBo7R0mMs061NReEG1VCrJUJtx+fP7TDs49UBrp7bi9bvi6XgiDMrTgrHlmSePtCN8PjXipE/dqKEAGbIAirRs1QPSUJRZjluRvcmmRToL3/YC2GYXCi5zTRZjvrk8qXbW2NXaMiHVIQCNSwQWB4dqj4/Brv1PSyvSwFm0U0uBaEcLFZTOSlx3Kqpg8Q9WsrRQRsgiCsCoPuYfrcA/OmQ06qSCpjyDNM81gbFwYusSNtC6Z5Ar3FGh73MjTmFemQgsBVAdtY6AK2s/UDuL0ae0WzEUEIu7LcQHv/+BhL2IbQ32hEwCYIwqpQM1wHQHlCcAHbusRAU4mf1vwKVVe5KX3u7pBL0dg5CiA6RAoCgRo2CO3w7CNV3STFWSkTT/MFIewmZxpWiPq1FSMCNkEQVoXqoXriLXFkzDJD7VpJUQmk21PpcvaQbk8lNzZ72dbW2DWKSZHITYtdtvcQhNXiyvDs0LT2Hx73crF5iJs3pIvhvIIQAUqzHSTEWtlRnhrupdwwRCK4IAgRTzd06oYa2JBcsaCneeuSyuhx9XFT+vZlfQrY2DVGXnosZpN4BiYIVotCtM0Usk6Rxy/1YBiwZ4PoDikIkcBuM/Evv7s33Mu4oYi7C0EQIl77eCdO1RV0/dqkXenbyY7JnHdY9lKomk5L97hoOCIIV0mItYakhs0wDI5W9lCUFSdqZQRBuGGJgE0QhIjmUb38vO5pTLKJisTSBZ2bE5vJn+36A+Kty1db1tY7garpouGIIFwlMc4Wkhq21t5xOgec7BW7a4Ig3MBEwCYIQsTya36+XfkD2sc7+cT6DxNriQn3kq4jGo4IwvUSYq0Mh6CG7UhlDyZFZleFqJURBOHGJQI2AU3XONj2Jm7VHe6lCMIUTdf43sUnqR1u4CPl72VzyvpwL2lGjV2jJMRaSbzcGU8QhEDANuby41f1RV9D1XROXOplW2kydps5hKsTBEFYXUTAJnBpqJbfNLzAgdZD4V6KIACBJiNP1vyK8wMXeU/Ju5e1Bm2pGjvHxO6aIFwjIQTDsy80DjLh9otmI4Ig3PBEwCbQMNIMwFsdR3H5XWFejXCjMwyDXzc8z/GeU7yr4G7uyLkl3Eua1ciEl8Exj6hfE4RrJMZensU2tvi0yCOV3cRHW1hfIGavCYJwYxMBm0DjSDMOazwezcuhjiPhXo5wg3up5SBvtB/m9uy9vCv/rnAvZ06NnWMAImAThGskxl3eYVtk45Fxl48LjYPcvD4dRRa3KoIg3NjEb8EbnE/z0Tbeyc60rWxMruBQ+xE8amhm5wjCQh1qP8ILza9yU/p2Hi95aFlnp4VCY9coiiyRlxZ5zVAEIZwcMUsL2M7WD6DpBrvXp4VyWYIgCKuSCNhucC1j7WiGRpEjn3vz7sSpujjcdTzcyxJuMMOeEX5Z9wy/rH+GTcnr+XD5e5ClyP/11Ng5enlgthLupQhCRImymoiyLn549qnaPlIdUeSkiochgiAIpnAvQAivxpFmJCSK4vOxm+2UJ5TwWttb3Jq1B4siunIJCzfgHqJy4BKlCUVkRqfPuUvW5xrg1dZDnOg5jYHBnoydvK/0ERQ5vAGQ26vyysk27tmZM2t3OlXTaekZ5/YtWSu8OkFYHRJjrYvaYXN5/FS3DHP3zpyI32UXBEFYCSJgu8E1jDSTEZ2G3WwH4N78/fz72f/mWPc73Ja9J8yrE1Ybv+bnvy98ny5nDwAOazzrEktZl1ROeWIxUaYoALomenil9XVO955HkRX2ZO7i7tzbSIpKDOfypxyt6uHZIy10Djj53CMbZrxpbO+bwK/qFGWJDpGCMJOEWCtDi2g6cq4hkA65vSxlGVYlCIKw+oiA7Qam6RrNY63clH6lZXqJo5DC+DxebT3ELZk3hX2nQ1hdnm16mS5nD09UvB/N0Lg4WMuZvkqOdr+DLMkUxucRFxXNme4qLIqF/bn7uDPnVuKtkRX0nKrpw6RInK7t59C5Lu7Yev0u2pWB2aLhyCTD5wLFgqSIPy1CIGBr75tY8Hmna/tJiLVSkBFZvxeEtckwDAzVH+5lCMKcxF/VG1jnRDdezUdRfP7Ua5IkcW/efr554Xuc7DnDzZk7Zz1fN3QuDdZSllCMWaRP3vCqB+t4vf1tbsveMzU3bU/mrssPBtq4OFjDpcFa+ocGuD//Lm7P2UuMOTrMq77eqNNHXfsID+7Jp6l7jJ+9Vk9JdjzZKdNraZq6xnDEWKa64d3oDJ8b5y/+HFP+Nmy3PBHu5QgRIDHOxpjTt6Dh2R6fSlXzELdtzkQW6ZDCClBr36b15M+Jeu9XkaPEQwIhMkV+Vb+wbBpGA/PXihwF015fn1ROTkwmB1rfQDdm/kM77pvgP859l29e+B4vthxc9rUKkW3C5+RH1T8nPTqNR4oemPY1RVYodhTwcNH9/NmuP+A7j/wjDxbeE5HBGsCZun4MYGdFKp96cB1RVhPfeuYiXr827biGzlGKMuNFjc1lvrPPYbhGUJtPYczye0O4sSTEWjFY2Cy2yqYh/Kou0iGFFeNvOonuceKvfTvcSxGEWYmA7QbWONJMki2BBJtj2uuSJHFv/p30uQc403fhuvOaRlv5+3f+nabRFrJiMsTA7RucYRg8WfMUTr+L31r3wYhqVnO+YYBv/OoCE+7g011O1fSRnmgnKzma+GgLn35wHV0DTn56sH7qmFGnj4FRMTB7kj7Wj6/yAFJ0IoZ7DL2/JdxLEiJAYmxg93lg1B30Oadr+4izmynJdizTqgThCsPvReuqAcBf/QaGLh42CZFJBGw3KMMwaBxpuW53bdLmlPWk21N5peX1qV02wzA41H6Er535Joqk8P9t/10+tu4DeDQvb4iB2zeso90nOT9wkYeK7iMnNjPcywFANwyePdzMvz91gbP1A7x5rjOo88ZcPmrbRthRnjK1c7a+IJH7d+fy1vku3qnpA6Bpsn5NNBwBwHvyFyDJRN33RZAk1Lbz4V6SEAESLgdsgyPB7bD5VY3zjYNsLU1BlsXOtbD8tK5q0FVit96NMT6A1nH9Q2pBiAQiYLtB9bkHGPdPUBw/c8AmSzL35u+ny9lD5UA1HtXL9y/9lF/WP8O6xDL+dOcXyInNIismg03J6znUfhi3uvBuYMLq1uvq56m6ZylLKGZ/zr5wLwcAl0flP35VydOHm7l5fTql2fEcOtuJFsST03P1A+iGwY6y1GmvP7qvkMLMOL7/Ug39I24apgZmxy7Xt7FqqD11qE3vYNl8P0pSDkpaCWrbuXAvS4gACbE2APpHgtthq2oewuvTRDqksGLU9gtgspJ018eQouLxXXw93EsShBmJgG0N6XcN8o9vf5NR79i8xzaOTNav5c96zPbUzSTbEnm+6RX+6fR/cLr3PA8V3sdnNn1sagwAwH35+3Gpbt7uOLbk70FYPTRd4/sXf4pJNvHEuvdHxKDrrgEnX/nhKS40DvKhu0r41IMV3L0zl8ExL+cbBuc9/1TNzMN6TYrMZ969HjD472cvUt8+Sm5aDBbzjd1F1TB0vMd+imR3YNn8LgCU3M3oA63ozuEwr04IN7vNhM2iMBhkSuTp2n6ibSbKcxOWeWWCEMgaUtsvYMpah2yJwlxxG1p7JfpYf7iXJgjXCf8dlhAShmHw87rfcKrrAq+3z1842zDSTIw5mjR76qzHKLLC3Xm30+XsYcI3wee3fIr78vdfd2OeF5fDusQyXmt/C6/mW/L3IqwOLzS/Stt4Bx8qfw8Oa/hruU7X9vGVH57C7fHzRx/cwl07AkN3t5QkkRhn5bXTHXOeP+H2U906zPar0iGvluKI4rfur6Cpa2yq4ciNTq0/ht7fjHXXe5HMgfQ3U+6WwNdEWqRAIC0ymBo2VdM5Vz/AluJkTIq4NRGWnz7ajTE+gJKzEQBz+e0gSfir3wjvwgRhBuK34hpxYeAS1UN1xFpjONJ1Ao/qnfP4xtEWiuLz5+1wd3PGTt5X+gh/uvP3KU8smfW4+/LvZMLv5Ejn8UWtX1hdGkaaOdD6BrszdrA1dWNY16LrBr9+q5H//E0VmUnR/NVv7aTsqif0iixzx9YsqluH6Rxwznqdc/WBYb3XpkNebWd5KrduDtTpFd7g9WuG34v3naeQUwowldw89bqckIkUm4wmAjaBQOORYGrYalqHcXlVtol0SGGFaG2VAJhyNgEgxyRiytuKv/ZtDFU8fBYiiwjY1gCf5udX9c+RHp3GH+39DG7Vw/HuU7MeP+odY8A9OGvDkaspssJt2Xuu6yR5rSJHPqWOIg62vYlfEwMo1zLd0PlF3dMk2By8t+Td4V4Ob1/o4vmjrdyyKYM//fBWEuNs1x2zb3MmJkXm9TOz77Kdqu0jKc5GfvrcdWkfuquEj95TyvbSG/vG0nfhJQznMNbdH0C6atddkiRMuZtROy+Kmx6BhFhbUDtsp+v6sVoUNhQkrsCqBCFQvyY7MpFjk6deM6/bj+EZR22e/R5qkmHoaMNdGLq6nMsUBEAEbGvCwbZDDHqGeF/Jw5SnFFMQl8cbHYdnnaHWEET92mLcl38no75xjnW/E9LrCpHlRM8ZOie6eaTofmym64OjlfbmuS6yU2L4+P3lmE0z15TF2S3cVJHK0coeXJ7r/7i6PCoXm4emdYecjcWscMe27Fnf60agO4fxnX8RU8EOTBll133dlLsFVN9Uu2zhxpUQa2V4zDNn0x9dNzhb18/moqQb+t+VsHIMvxetuxYld9O015WsCqT4dHyX5m4+YhgG3rd/iOuXf87E9z+H67m/x3vyKdS28xje2TM5BGGxRMC2yg26hzjQ+gZbUzdRllgMwP7cfQy4B6kcuDTjOY2jzVhkMzkxWSFdS2lCEYXxeRxoPYQqnjitST7Nx/NNr5AXl8O21M3hXg5tveO09Iyzb3PGvIHWnTuy8fo1jlR1X/e1cw3986ZDCld433kKdB3rTe+b8etKRhmYrKJbpEBinBXdgNGJ2Xdb6ztGGHP52S7+/QkrROu6BLo6lQ45SZJkLOvuQO9tQBtonfV83/kX8dccwly2D3P5bRiqF9/5F3G//DUmfvC7OH/553hP/ELsvgkhIwK2Ve7XDc8jIfFY8QNTr21OXk+iLWHW5iMNI80UxOehyME9yTQMI6jjJEnivvw7GfaOcLLnTFDnCKvL6+1vM+Id5bHiB+cNkFbC2xe6MSkSN69Pn/fY/PQ4ijLjeP1MJ/o1P9OnavpJiLVSkHlj16UFQ+tvQa07gmXjPchxM99gSyYLpuz1gafNQf7+ENampMspym+c7UTXZ/5ZOFXbj9kks7FQpEMKK0NtrwSTFSX9+tp8c+ktoFjwX5q5+Yi/8SS+k7/EVLgL660fx7bnw0Q/+tfE/NY3iXrwT7DseCwwIuD8i2jtVSFbsz7ai/OXf4422B6yawqrhwjYVrGaoXrO9Vdxb/5+Em1XN1lQuD17Lw0jzbSNTa/ZcatuuiZ6gqpfA+gbcfO5f32LuvaRoI5fl1hGbmwWr7S+gaZrQX8vQuQb841zoPUNNqdsoDjIn5/l5Fc1jl/sYVtpCjFR5qDO2b89m94hF5dahqZec3tVqpqH2FGWihwBQWgkM3QNz9EfI9lisWx9aM5jldzNGBOD6MNzd+cU1rbyvAT2bcnihWOt/OOTZxgcnd6ARDcMztT1s6EgEZvFFKZVCjeSq9v5S8r1fzskazTm4pvwNxzF8LmmfU3tqcdz6NsoaSXYbv/U9PpdsxVTZgXWbe8m6v4/BLMNtfVsyNbtrz+KPtyF9+iPxYOwG5AI2FYpVVf5Rd0zJNsSuTPn1uu+vidzJzbFet0uW9NoKwbGrAOzr3Wmth+vX6M2yIBtcpdtwD3I6T7RJW4tebH5IH5d5eGi+8O9FADO1A3g9Kjs25QZ9Dk7y1OJi7bw2qkrQcT5xgFUTWdH+Y3dRCQY3uM/R+9twHrzB5EsUXMea8oNpMyqreL3wI3MpMj80Ue286kHK2jrm+Cv/vckJ6t7p77e3DXG8LhXDMsWVsy17fxnYl53J6g+/HVHrjqvB88r/44Uk4Tt3i8gmSyzni8pZkw5G1Fbz2HM0k9godSW02CyoHXXBtUURVhbRMC2Sh3qOEKvq4/3lL4b8wxPiKJMUdycuZPTfecZ8Y5Ovd4w0owsyeTH5wb1PucbBoDAQOJgbUxeR2Z0Oi+3vD5r4xNhdelx9nKk6wT7snaTZo+MG6u3L3SRFGejIj/4IbsmRea2zZlcaBykbyTQue50TT/xMRaKssRctbn4a9/GX3UA84a7MZfsmfd42e5ATikQdWwCkiSxZ0MGX/7ELjKT7HzrmYt89/lLuL0qp+v6UWSJLcXJ819IEELg2nb+M1FS8pFTCvFfeh3DMNA947he+hoA9vv+ENk2dzdhAFPeVgz3KHp/85LXrI/1oQ91YN3+KHJiDt4TPxddeG8wImBbhUa9Y7zUfJD1SeVsTF4363G3Z9+CYRi82XF06rXGkWZyYrOwKrM/GZrk9Pip7wgEe539E0GvT5Zk7svfT6+rj4uDokvcWvB044tYZAv3598V7qUA0D/i5lLLMPs2ZSw4jfH2rVnIssShM514fCoXmgbZUSrSIeei9TbgefsHKFnrse7+QNDnmXI3o/c2onvGl3F1wmqR6ojiTz+yjXfvzefYxR7++n9PcvxiDxX5CdhtwaU1C8JSqe0XkBOmt/OfiWX9fvSRbrT2SjyvfB3DOUjUvb+PHJ8W1PuYcjeDJKO2LD0tUm0+Hbhm4Q6sez6EMT6Ar/KVJV9XWD1EwLYKPd34Iqqu8p6SuWtIkqMS2ZyygcOdx/FqPvyan9ax9qDTISubBtENg9IcBz1DrjnbMl9rXVKg1XefayDoc4TIVDfcSOVANffm3UGsJSbcywHg8IVuJGDvxowFn5sQa2VbaQpvX+jidG0/flWkQ85FnxjCfeDrSDGJRN35O0hBNiuCy+39MdDaLizb+oTVRZFlHtlXyJ9+eBsAIxM+0Z1VWDGG3xNo5z/H7tokU+EusEbjfvUbaL312G7/7RmblMxGskajZJShti69CZu/5TRyUh5ybAqmzApM+dvxnX0e3Tm85GsLq4MI2FaZ+uEmTvacYX/uraQGkZq2P2cfLtXNie5TtI53oBpa0A1HLjQMEms3c8vGDFTNoG94/uGnk2yKDVmScfld8x8sRCzd0PlNw/MkWB3cnnNLuJcDBGY2Ha7sZn1BIknxi5sDd+f2bJwelZ8erCfObqYk2xHaRa4RhurDfeDrGKqPqHt/H8m2sIBdTs5FsjtQ2+auY1O7a9FHepayVGGVKcl28OWP7+IT76pgz4b5u7wKQihoXdUztvOfiWSyYC7bB5ofy673YS7ateD3M+VtRR/uQh/tnf/gWeiuEfTeBkwF26Zes+5+P+haYMSKcEMQAdsq4lE9/Kj6FyTbErkv/86gzimMzyMvLoc32g/TMNIEQFF8/rznabpOZdMgmwqTyEkN3KR19gdfxyZJEnZTFBOqCNhWs1O952gb7+TdRfdhmaFWMhwutgwxPO5l3+bgm41cqyQ7nuyUGFxelW1lqciySIe8lmEYeN78X/SBVqL2fwYlYeFzGyVJxpS7CbW9EkO7fh6RYRh4zzyD+7mv4j707VAsW1hF7DYTt2zKwKSIWxFhZczVzn8m1h2PEfXAH2PZvLhmW6a8rYH3XUK3SLUlsENnyt8+9Zocl4pl4z2odUfQ+poWfW1h9RC/JVeRX9U/z5BnmCfWfSCoGjQIBE77c/bR5x7g9fa3SbenEmOJnve8xs4xnB6VzcXJpCfZkYDOBTQeAYg228UO2yrm1/w82/gyObFZ7EjbEu7lTHn7fBcxUeYlNSmQJIm7dmQDgc6RwvV8519EbTyOZefjUzcdi6HkbgG/G62nbtrrhubH88a38Z36DVJMEnpfE/rE4BJXLQiCMDPDMFDbzs/azn8mkskSOH6RNc5yXApyYvaSAzYpLg35modmlq0PIUXF4Tn2pGjzfwMQAdsqUTlwiaPdJ7kr9zaKHPkLOndrykYSrA6cflfQ6ZDnGgZQZIn1BYlYzQopCVELajwCYDfZcfmDT6MUIsuBtkMMe0d4tOgBZCkyflWMuXycrR9gz4Z0zKalremWTRl86YntVOQF32XyRqG2nsN38ilMRTdh2fLAkq5lyloHimlaWqTuGcf9wj+hNhzDsuNR7O/6o8D7NolW1YIgLA99pBtjYjCo+rVQMuVtReupw/As7B4KwPA60TqrMRdsvy5olCxRWHe+B723AbXxRKiWK0SoyLgLE+Y04XPyk5qnyIrJ4IHCexZ8viIr3JYdaMMd7MDj8w0DlOU6iLIGBplmJUcvYoctCqdIiVyVqgfreKn5IDvStlCWWBzu5Uw5XtWDphvs27TwZiPXkiWJokzRyv9a+ng/7tf/Gzk5F9ttn1j0k+VJktmGklkx1d5fG+nC9Zu/Retvwnbn72Dd9jCyIx05KQd/8zsh+A4EQRCup7UHmh+Zclc+YOPy7t5CqW3nwdAw5W+b8eumsluQk/PwnvgFhupd6lKFCCYCtghnGAY/rf01Lr+bj637AGbZtKjr3Jq9h3cX3seWlA3zHts37KJ70MXmoispZ5nJ0fQNu/GrwXeKjDZHi5TIVWjAPcT3Lj5JRnQaHyp/T7iXM8UwDN660E1hZhxZKZHRrXKtMXQdzxvfAQyi7vo8kskakuuacjdjjPbiu/Q6rqe/AqoX+4N/irnopivHFOxE720QXc8EQVgWantloJ1/TNKKvq+ckh9ovrSItEi1+TSS3YGcWjjj1yVJxnrzhzCcQ/jOv7TUpQoRTARsEe6d3rOc66/kwcJ7yIpZ/K6CVbFwb/5+LEHUvp1vCNSRbC6+8kstKyUaTTfoHQo+ALObo3CKgG1V8Wk+vlP5Q3QMPr3xiaBrJVdCU9cYXQPOkOyuCTPznX8RracO296PIseFbtSBKXczAN7DP0SOTsT+yF+ipE3fuTUX7gRAbRZpkYIghNZC2vmHmiTJmPK2BpovLWDYtaF6UTsqMeVvQ5qjLMGUUYapcBe+cy/iOfZTfFUHUdvOoQ13ieHaa8jitmuEFTHsGeEXdU9TGJ/PXbm3rdj7nm8cICPJTmqCfeq1rOTLnSIHnGSnBre7EW2y49G8aLqGsoDZTUJ4TO7mdk5089lNv0WqffFNPZbD2xe6sJhldlUEN7RUWBitvxnfqd9gKtyFqWRPSK8tx6ag5GxCkhVsd3wayWK//hhHBnJCNmrTO1g23B3S9xcE4camdQbfzn85mPK24q9+A62rJuiUTLWjClTftO6Qs7Hufj/6xCD+6jfgmiBNsjuQ49Ow3fZJ5DjRZGu1EgFbhNINnR9V/wLN0Hmi4v0r1vTB7VWpbRvh7p05015PT7QjSxKdAxNAcDfMdnPgpsyluiNm4LIwuzc7j3Ky5wwPFNzNhuSKcC9nGo9P5UR1HzvLU6fqKoX5GbqOJM//u8Pwe3G//t9I9nhs+z625Lq1mdjv/8N5jzEV7sR3+ml01wiy3RHyNQiCcOMxDB3v+ReQbLEo6aVhWYOSVQFmG2rrmeADtuYzYI1GySyb91g5JonoR/4SwzAw3GMY4/3o4/3oY4H/U+sO4284jnXbu5f6rQhhIlIiI9RbnceoHW7g8eIHSbGvXL71xeYhNN1gc9H09zSbZNISoxY0iy3aFAUg0iJXgYaRZn5V/xwbkyuCnvG3kl473YHXp7Fv0+Jnr91oDNWL8ydfxPXSv8xbF+Y9/jOM0d7A7pd1/rEfy8VUuAMwRFqkIAgh469+E723Aevu9yMp4XngJylmTNkbUFvPYRjz9wIwdBW17Rym3M1IC+hdIEkSsj0eJa0Yc/HNWLe9m6jbP4mcnIfWUbWUb0EIMxGwRaA+1wBPN7zIuqQy9mbeNP8JIXS+YYBom4ni7Ou752UmR9O1gE6R0ebAjZ9LdIqMaCPeUb5b9SOSbYl8bN0HIqaF/6TatmF+81Yz20tTKJnh51KYmdbXjOEeRWuvwvnUX+BvmrkDo9p6Fn/1G5g33YcpM7w7q0pCFnJCJuosaxUEQVgI3TWK9+QvUTLKMZXsDetaTHlbMVwj6P0t8x6rddeB1xlUOmRQ7529Aa23AcMn7sdWq8i6MxMAeKH5AJIk8eHy9yxLatJsdN3gQtMgGwuTUGZIo8q63CnS59eCup7dLHbYIp2qq3y38sd4NR+f3vgEUZd3RSPFyISXbz1zkZSEKD7xQMWK/ntY7SYHVdsf/hJyXCqeg/+J+9B3MHxXZiPqrlE8b/4vclIu1p2PhWup05gKdqJ116G7RsO9FEEQVjnv8Z+C6lu2VO+FMOVuBkkOqluk2nwKFAumnPk7ewdDydkIho7aWR2S6wmh5699e86vi4AtwvQ4+zjde57bsvbgsK7sbkJz9xjjLj+bimdOwcxKicEAugeDC8CiL9ewiYAt8rhVD0e6TvAvp/+T5rFWPlrxPjJj0sO9rGlUTeebT1fh9qn87qMbRO3aAmm9DcgJmShpxdgf/hKWbe9GrT+K86m/QO2uxTAMPG/+D4bfg23/Z5AUc7iXDATq2MBAbTkd7qUIgrCKqR1VqA3HsWx5ANkR/u7Cki0GJb1k3oDNMHTUljOYcjaGbLSKklYEZptIi4xQ/roj+M48M+cx4g4owrzUchCzbOLO3FtX/L3PNw4gSxIbC2cO2DKTAymOXQNO8tJj572e3XS56YgI2CKCYRg0jrZwrOsdzvSdx6f7Sben8pGK97EtNTyds+by1KFG6jtG+e2H1pEt5q4tiGHoaL31mAsDKdWSbMK64zFMOZtwv/Ft3M/9PUr2erSOKqx7PoKSkBXmFV8hJ2Qhx6cHukWu2x/u5QiCsAoZqg/P4R8ixadh2fJAuJczxZS3De/xn6KP9c86OkXvb8Zwjcw6LHsxJNmEKbMCtaMKwzDCvtsoXKH1NuB563tYinfOeVxQAdtzzz3HN7/5TVRV5WMf+xgf/vCHp3394MGDfOMb38AwDLKzs/nqV79KfLyoNVmoyd21u3JvC0tXxXP1g5RkxxNtm/lJe1pCFIos0TEwEdT1bCYrEhJO1T3/wcKycfndHO46zrHud+hzDWBTrOxM38rNGTvJj8uNyF/cp2r6OPBOO3duy2b3+sja+VsN9OFO8LlR0kumva6kFRP9+N/iPfZT/DVvouRsxLw+sprMSJIU6BZ57nl09xhyVFy4lyQIwirjO/scxlgfUQ/8MZIpcuaJmvK34j3+U9TWM1g23jvjMWrzaZAUTHlbQvreSvYG1NazGGO9SPHi72ok0CcGcR/4OlJMIra9T8x57Lwpkb29vXzta1/jySef5Omnn+bnP/85DQ0NU1+fmJjgy1/+Mt/+9rd59tlnKSsr4xvf+MbSv4sb0Mstr4Vtd21w1ENH/wSbi2efvWVSZNKT7HQF2SlSlmSizXaREhlm37v4JM80vkSsOZaPVryP/3fLX/Kh8vdQEJ8XkcFa96CT/3mxmqLMON5/Z/H8JwjX0XrqAa4L2AAksw3brR/H/tjfEHXX5yPyZ8BUuBMMA7XlTLiXIgjCKqMNd+E7/yKmkj2YstaFeznTyHGpyAlZqK3nZvy6YRj4W06jZJaHvGOvKWcjAGq7SIuMBIbfi/uVr2OofqLu/QMk6/XzSa827w7b0aNH2b17Nw6HA4B7772Xl19+mc9//vMA+P1+/vqv/5q0tMBsrrKyMp577rklfhs3nl5nH6d6z3Fn7q1h2V270DgAwOZZ6tcmZSVH09Q1FvR17eYokRIZRm1jHVwaquWhwvu4Lz/y08s8PpX//E0VFpPM7zyyAZMiymwXQ+upR4qKR4qdOeUGQEnOW8EVLYycmIMUnxZIi6y4PdzLEQRhlTAMA+/hH4DZhnX3B8K9nBmZ8rbiO/8irpf+FQwdDOPK/9c1jNFeTLPsvi2FHJeKFJeK2lGJZcNdIb++EDzD0PEc+g76UBtR934RJWH+kUXzBmx9fX2kpFz5o5+amsqFCxem/ndCQgJ33303AB6Ph29/+9t89KMfXdDCk5JEfcrPGp/Coph5/9Z3EW+bvz5sLikpCz+/un2UjKRoNpalzfnEvTQvkZPVfcTERQXVBCI+Kha/5FvUmiLJal3/j+oPE2W28fjme7BbIqcD5Eyfp2EY/POPT9Mz6ORvP7OHsqLZgw3held/pm39jdjzKkhNXb3phEPr9zJy7GkSow0U+8p/H6v133ykEp9n6InP9Hrj519noruW5Ad+h7jchdXmrtTn6b/5XvoG6kF1gSSBJCMpMkgySArWkp2k7NqPEhX69UjFWxmvfJPkRNuKNJoSP6MzG3rr56jNp0i862M4tgc3bmLeO25d16fdwM9WrDg+Ps7v/u7vUl5ezqOPPrqAZcPg4AS6bizonLWk19XP4dZ3uDP3VnzjEv3j44u+VkpKLP39Czt/zOnjXF0/d2zNYuCa+jTDM4H7wNex3vJRlMQc4qMC/8Ara3spyJj/JsqClRHX2ILXFEkW85lGgl5XPyfaz3J33u04R1WcRMb3cPXn6Vc16jtGudQyzMXmIVp7x3n8tkIyHbZV+ZmHy9Wfqe4cRh3tQ1l356r+DLX0zWD8mp7Tb2Epv21F33u1/puPVOLzDD3xmV5Pd4/hfPX7KOmleDJ34l3A57Oyn2cslnf92ZxHDE0AE6Ffjz+5DMP/Cj1VZ5d97qb4GZ2Zv/Eknrd/gal0H76C26c+I1mW5tzAmjdgS09P59SpU1P/u7+/n9TU1GnH9PX18clPfpLdu3fz53/+54v8Fm5ck7Vrd+Wu7E2Jy6Pyysk2DpxqR9cNblqXdt0xakclWk8d/upDKHs/SnZKIKe6o38iqIAt2mynx9kX8rUL8zvYegiTrHBHzi3hXsoUwzBo7BjhyNkOLrUMUdcxil/VUWSJosw4PrC/mLt25oR7mavaVP1a2vX1a6uJnJSLFJsSSItc4YBNEITVxVC9eF77Jvg8WG/5GJIk0ulnYsqsAElB66ha9oBNuJ7W34Ln0HdR0kqw7XtiQTXk8wZse/bs4Rvf+AZDQ0NERUVx4MABvvKVr1x5c03js5/9LPfffz+f+9znFvcd3MB6Xf2803OW/bn7Vqx2zevTOHi6nZdPtOH0qOwoT+WRWwqm2vZfTeuqAQJdi4w9HybFEYXZJNM1EFzjkWiTaDoSDiPeUU70nGFv5i7iLJGTkvDUm428dLwNCNRD3r4li3X5CZTlOrBZxJSRUNB66sBkQU7ODfdSlkSSJMyFO/FdeAXDM4FkE6nzgiBcz1B9uF/5OlpXDbY7Po2SGDljSiKNZIlCSS9Gba/Cuuu94V7ODcXQdTyHvo1ki8F2z+8tOCV13juktLQ0vvjFL/LEE0/g9/t5z3vew6ZNm/j0pz/NF77wBXp6erh06RKapvHKK68AsGHDBv7u7/5ucd/RDeaVltcxySbuzr192d/Lr2ocOtvFC8daGHP52VSUxKP7CuecqaZ214DJiuEaQe9tREkvISPJTmeQAZvdHIVH86DpGoqshOpbEebxWttbGBjcucK7tnM51zDAS8fbuHNnDg/clIsjJjQDQYXptN56lNQiJHn1B8Cmwp34zr+I2noWc9m+cC9HEIQIY2h+3K/+B1rnJWy3fQJzyZ5wLyniKdkb8L3zKzE2ZYWpDUfRh7uw3fW7i/rcg/qL/tBDD/HQQw9Ne+073/kOABs3bqSmpmbBbyxAn6ufkz1n2J+z/LtrlU2D/ODlGobGvJTnOvj8bUUUZ809K093DmOM9mLZ9jC+cy/gbz6Fkl5CVnI0NW0jQb2v3Xx5eLbqDkv3yxvRhN/J4a4TbE/dTHJUYriXA8DQmIf/ef4SuakxfO7xzYyOiF3X5WD43OiDbVi2PBjupYSEnJyPFJuC9+zzKJnlyHN0vRQE4cZiaCruV/8Trf0C1n2/JR7qBMl0OWDTOqqQRYC7IgzNj/fUb5CT8zEV7FjUNUSSbxi9fHl37a685dsF8asaTx6s42u/OE+UxcT/+cAW/vhD2+YN1gC07logMOhRyV6P2vQOhmGQmRzN8LgXl0ed9xrRpssBm0iLXDFvdRzFp/m4J++OcC8FAFXT+dYzF1F1g995ZAMWs9hpXS5aXxMYBkp6abiXEhKSJGG7/VMYnnFcT/9ftP6WcC9JEIQIYOgqnte+idZ2Duvej4rxHwsgJ+ch2WJRO1bXPDa1vRJ9YjDcy1gUf/UhjIlBrLves+jZpyJgC5OuiR7e6T3LrVk3L1uNUUffBH/7g1McPNXBXduz+cuP7WBdfvA7LlpXDViikBNzMRfuxHAOofc3k5US2CkLpo4t+vIOm1N1L+6bEBbEq/k41HGEjckVZMakh3s5ADz9djMNnaN87L4y0hLnHgwpLI3WWw9IKGlF4V5KyJgyyrA//CVQTLie+ypq2/lwL0kQhDAydA3P699GbTmN9eYPYVl/Z7iXtKpIkoyStR6towrD0MO9nKD4m0/jfulf8J1dfXOeDb8H35lnUTIrULLWL/o6ImALg7rhBr525pvYTVHLsrtmGAavnmrnb39winGXnz9472Y+dHfpgnc2tO4alPRSJFnGlLcVJAW1+RRZl5uTdF4zAmAmUwGbP7iaN2FpjnSdwOl3RczuWmXTIC8eb+XWzZnsXhcZAeRapvXUIydlI1nWVmCsJGRhf+QvkePTcb/y7/iqD4V7SYIghEGgccN3UZtOYr3p/Vg23hPuJa1KppwNGO4x9MH2cC9lXtpIF55DgTIsfagzzKtZOF/lAQzPONadjy96dw1EwLbijnSd4BvnvkucNY4/2vH5kO+uDY95+Novz/PTg/Wsz0/gbz+xi01FSQu+ju4aQR/twZRZDoBkjUbJqsDf9A6JcVasZiWoxiP2qZRIscO23FRd5bW2tyh2FFAYnx/u5TA87uU7z10iOyWaD921ulvMh4thGKidl/C8+b9ow11zH6traH2Nq76d/2xkuwP7Q3+Kkr0e79vfx/vOrzCM6+d3GoaONtKFv+5IIEVUEIRVzzB01NazuJ79O9SGY1h2Po5l8/3hXtaqNbnTo3ZcDPNK5mb43HgOfANJMaPkbkYb7pzx936kMjwT+M6/hClvK0pa8ZKutfrbiK0SuqHzm4YXeL39bSoSS/nkhg8TZYoK6Xv0Dbv46k/O4PKofOSeUu7YmrXoaH6ynb+SUT71mqlwJ963vgdD7WQm2+nsDyYlMvA9OlVRw7bc3uk5y4h3lA+VvyfcS0HTdf772Yv4VV3UrS2CofpQG0/gqzyAPhR4AqqP9RL14J/O+m9aH2oHv2fN1K/NRLJEEXXv7+N9+4f4zj6HPjGIddd70Qda0foa0fqa0PqbwBd4QCTFphD9gX9c0lNNQRDCx9C1wO/Ccy+iD3cgxSZju+2TosHIEsnRCciJ2WgdlbDlXeFezowMw8Bz6Lvoo71EPfBH6EMdaG3nMdyjSHZHuJcXFO+5F8DvwbLz8SVfSwRsK8CjevjexZ9SNVjNbdl7ebz4wWVpcf/i8TbcHpW//NgOslOW1pFR664BcxRy0pVZTqb8bXjf/gFq0ztkJq+jqmlo3uvYTDYkpDXZdMSvq8hIETGuQDd0Xm07RHZMJusSw3/D/szhFuraR/jUgxVkJF0/30+Yme4awX/pDfyXXsfwjCMnZmO79RMYPife4z9Haz+PKXfLjOdODcxOX5s7bJMk2YT11o8jxSbhO/Ub1Pqjl78gISdmYy68CSW1EH28H9/Z5zBGe5AcGeFdtCAIC2KoPvy1b+O78BLG+AByQha2O34bU9FNSBHwN3ctULI34K86iOH3Ipkjb8yO7/yLgTrF3R/AlFmBenlnTR/qRF4FAZvuHMZ/8SCmkptRErOXfD0RsC2zQfcw37rwPXpcfby/9BFuzV6eFqpjLh9Hq3q4c2fOkoM1CHSIVDJKp/1ilG2xKJnl+JtPkZW3kyOVPUy4/cREzT78T5Zk7OYonGssJbJ+uIn/qfoxeXE5fHbTb4X9Cf6F/ov0uvr5xPoPhX0ttW3DvHC0hb0b09mzQdwoB0N3jeA9+RRqwzHQdZTczVg23oOSWYEkSRi6iq/6EN4Tv0TJ3oQkX5/NrvXUI0UnIscsPAV6tZEkCeu2h1GS89CHu5FTC1GS86fddOjjA/jOPofadgGLCNgEYdF05zBofuS41BV5P7X1HJ63/hfDPYacVoxtz4dRcjcjSaKKJ5RM2RvxX3gZrbt61geB4aJ2VOF75ylMhbswb7wXADkhEwB9uBOyF9+8Y6X4zjwDho51+6MhuZ4I2JZRr7OPr535Fqqh8rnNn6BiGXc+3jjTiarpPHzr0rvD6a4R9JFurDOkHJgKduA9/EPy7eMAdPZPUJabMOf1ok32NdN0xDAM3uw4yq8ansMiW6garKZqsJqNyevCtiZN13i++QApUUlsTd0UtnVMeu10B7HRFj5yd1m4lxLxDMNArT+K59iToHoxV9yBZcNdyPHTG7RIsgnrzvfgOfifqHWHMZffet11tN76NZ0OORNT7haY5UZDjk1GTshCbb+AZdO9K7ouQVhL3K/8O/pwB7a9T1z3uyfUDM8EnkPfRbI7sN35OZSMsrA/hFyrlPQSUCyoHRcjKmDTx/vxvPYtZEcWtts+OfXfX4qKB2s0+jz13JFAH+3BX/MW5nV3IMeFZn6oeFyxjN7uPI5b8/B/tv/usgZrPr/G62c62FSURE7a0puYTM5fu7p+bZIpfxsgkT5eDQTX2t9utuNaA239fZqfH1X/gl/WP8P6pDL+5uY/Ic2ewq/rn0fV559Jt1ze7DxKt7OXR4sfRA7zE0i/qlHZNMS2kmSsFpG2Mhd9Ygj3y1/Dc+g7KI5Moh//Cra9H7kuWJtkKtiBnFqI99SvMVTvtK+pY/0YzuE123BksZScTWjdNRh+T0iuZ2hqyK4lCKuBNtSOPtCCZI3B89b/4nn7Bxja8v298558CsPnwrb/M5gyy0WwtowkkwUlswytvTLcS5liqD7cr/4HhqERdc/vTcuakCQJJSErsMMW4bynfgOKCcvWd4fsmiJgWyaGYVA5cImyhGLSo9OW9b2OX+pl3OXn3l258x8cBK27Fsw25OS8674m2x0oGaWYu84RZVXoCCpgi8K5ymvYhj0jfO3Mf3Gi5zTvKrib3974MWIs0Txe8hB97gEOdRwJy7rGfOO80PQqFYmlbArjLt+k6tZhvH6NraWheaK0FhmGga/mTZy//BJaVw3Wmz9E1EN/hjxP2p4kSVhvej+GawRf5avTvuZpv9wkaI3Xry2UKXcT6Bpq56WQXM975Me4nv5KSK4lCKuBv+4ISAr2x76MedP9+KvfwPX83wfSJENM623AX/Mm5g13oyTlhPz6wvVM2RvQR3vQxwdW7D0NrxNtoAW1qxq15Sz+uiP4Lh7Ee/Z53Ae+jj7QStT+zyDHX3/vLCdkoo10RXSnSG2gFbXxBJYN9yDb40N2XZESuUx6Xf0MeIa4Mzf0c9auphsGr5xsIzcthvJcR0iuqXVNzl+beYfEVLAD79GfsD7RR1cwnSJNdvqc/SFZWzhc7KvjX975Nqqu8ZmNH2NTypXc6fVJ5axPKuel5te4KX07sZal1w8uxDONL+HX/by35N0R8STyTN0ANotC+Txpsjcqfbwfz1vfR+u8iJJRju22TyyoLsSUUYYpbyu+c89jLr8VOSoOAG97TeAhS6K4ybmakl4CZhta2wXM+duWdC1D1/A3vwNeJ7prZFUUvQvCUhi6hlp/FFPeZmS7A9vu96OkFOB587u4fv1lbHd/HlOIHhIZuobn8I+Q7PFYtz8SkmsK8zPlbMZ77Kd4Dv+QqLs/j2SyLOv76a4RnL/8EnhnuXdUTFhvet+sKZpyQhZ4D0Vsp0hD9eE98mOwRod87ITYYVsmVYOBlMENydenFYb0fZoG6R50ce/O3JDcsOuuUfSRrhnTISeZCnYAsNXWTueAc94nHdFm+6pt63+k8wRfOfTvRJuj+eMdn58WrE16vPhBfLqP55peXtG1NY+2crz7FHfk3EJa9MoUg89F1w3O1fezsTAJs0n8armWPt6P86m/QutrxHrLE0Q9+MeLKuK37HovqF58Z5+bes3TUY2SVjxjM5IbmSSbMGWtR22/sOQnslpP/dRNhtbbGIrlCUJE0zqqMNxjmEr3Tr1mLtqF/ZG/ArMV9/N/j+/S6yHZ7fBfeh19sBXrng8hWUI78kiYnexIx7rvt9DaL+A+8HUM1bes7+c79yL43Nju+G2iHvwT7I9+mej3/wPRH/06MZ/4NrGf/C6WzbOPGZATsoDIHKBt6Bqe17+F1luPbe9Hkayh7ZAtdtiWSdVANVkxGSTalnen4ZWT7STEWtlZEZob9sn6tcmB2TORoxOQ04opGKtjwp3PmMtPfPTsT2XsZjtu1YOmaxHRAj9Yfs3PU/XPUpFSzMfLP0KUyTbjcWnRqdyevZc32g+zL2sPObGZy7423dD5Rd0zxFtiuT//zmV/v2A0dY0x5vKztTQ53EuJSL4Lr4DqI/q9f4fsmLlOLRhKQibmstvwX3ody4a7kazR+PrasYin0jNScjehtpxGH+5AWcIOpNp6FmQTYKD1NmAu2B66RQpCBPLXHUGyxmDK2TztdSUxm+hH/xr36/+N9/AP8Ve/CYoCujb1f4aug64GRpPc9smpbICZ6K4RvO/8GiV7A6aCncv9bQnXsFTcjiTJeN76Hu6Xv0bUvX+wLG3+decw/urXMZfuxVyyuI7pU50iR7oiqlOkYRiBsVctZ7Du+TDm4t0hfw/xOHYZuPwuGkdb2JhUsazv09Y7TnXrMHdtz8akhOY/ZWD+2sz1a1czF+wg2t1DsjxGe9/4nMdGm+wAuNXVVaxfP9KET/fzUNndswZrk+7Pv4tos52n6p9ZkdzqY93v0DbewSPFD2CbZ20r5Wx9P4ossalQBGzXMjwT+GvfxlS8e0nB2iTLjkdAVvCefOrybo8h6tdmYcoJdE5V2y4s+hqGYaC2nkXJWoecnIfeJ3bYhLXN8DpRW85gKt6NpFz/bF+yRhN13x9g2fkeJKsdyWJHsjuQ41KRE7NRUgtQ0kvROi/h+s3foA20zvpe3uM/B82Pbe9HIiK1/0ZkLr8V2x2fRuuuwf3yvy5LcyXfuedBN5bUiGOqU2SE7bD53nkKf+1bWLY+hGXD3cvyHiJgWwaXhurQDZ0NycsbsL1ysh2rWeHWLaHb0dG6a1DSS+YdTDmZFrnT3sEPX65laGz2f9x2cyC9YbWlRV4crMEsm1ifOn+HT7s5igcL76VhpJmz/cvbccnld/Fs48sUxeezM23rsr5XsAzD4ExdP+V5CdhtYuP+Wr7qQ6B6Q9ZeXrY7sGy6D7XpJL6Lr4Iko6QufaTHWiRHJyAn5aK1Lz5g00e6MMb6MOVtQUktQutvwQhjZ1hBWG7+xhOgq5hLb5n1GEmSsW59EPuDf4L9Xf8H+31fJOqeLxB11+8Stf+zRO3/DPZ3fwkMA9czf4e/8eR111C7qlEbjmHZ8sCsHXKFlWEu2YPtjs+g9dTjfvFfMHyh6+6tTwzir34Tc9ktS2pzP9UpciRyWvv7LryM79wLmCtux7LjsWV7HxGwLYOqgWpizNHkxS1fA4DhcS8nq3vZtymDaNvsg6sXQnePoQ/PXb82SY5NRk4p4M7UPibcfv7pZ+cYnfDOeGy0ObDD5lpFnSINw6BqsIbShGKsQRbh7s3cRVZMBr9peAGf5l+2tT3ffACn38V7Sx+JmKeR3YMueofdbC0Ru2vXMjQ//osHUbLWoySFppMrgGXT/Ui2WLT2SixpBcuSwrJWmHI2ofXUY8xW6D4PtfVs4Dp5W1HSikHzoQ92hHKJghBR/HVHkBOy5s22mY+Sko/90b9GSc7D89p/Bdr2GzoQGJPhPfxDpNgULFseCMWyhSUyF+/GdufvoPU14Xrxnxb9O/NavrPPAwaWrQ8t+VpyQibacGdEdIr01x3Be/xnmAp2YN37xLLek4mALcQ0XePiYA3rk8qXdSbWa6c70A2Du3aGLigMpn7taqaCnZhH2vg/D2YzMu7ln392jnHX9QWrkwHbamrt3+ceYMA9yPqk4JvGyJLMe0rezZBnmNfa3lqWdXVOdPNWxzH2Ze1ekVq5YJ2tD3QB3VIsArZrqQ3HMVwjWDbdF9LrSpYoLNsfBsCWs7zNjVY7JXczGDpq58VFna+2nkNOKUCOTggEbARakAvCWqSPdKP3NWIu3RuSG1DZHk/Ug3+MufxWfOeex/3K1zF8bnyVL6OPdAcaNCxzd0IheObCndju/l30gVZcL/4zhm9p9276eD/+2rcwl9+GHLv0e4RAp0gnhnt0yddaCrXtHJ43/wclswLb/s8se9MvEbCFWPNYGy7VvazpkB6fyqGznWwrTSHVEbpuSlpXDZisQT9RMxfuBEkhrfKH/MFDefSNuPmXn53D6Zm+u2Q3rb6A7eLA5S6fCwjYAEoTitiSspEDra8z4g3tLxPDMPhF3dNT6ZeR5Gz9AAUZsSTGRUY9XaQwDAPfhVeQE7JRsjeE/Prmitsxb7ibuC2R0XgmUimphWCNXlQdm+4aRe9txJS3BQApOhHJ7kBbQB2b2nYBz9GfRMQTYUGYT2D2moRpkY0hZiIpZqz7Po5170fQ2i/gevpv8Z1+FlP+9sC8RCGimPO3EXX376H3N+OvWdoD6EBHYwnLlgdDsrapTpHDS0uLNAwDw+de1O9ltace96v/hZyUS9Q9X0BSQpPpNhcRsIVY1UA1siRTkTh/3dNiHb7QjcurhmxQ9qQr9WvB1SDJcSlE3fv76KM9ZJ36Bl+8N5WuQSf/+vPzuL1X6jumUiLV0OVDL7eqwRrSo9NIikpc8LmPFj+AjsGLzQdDuqbTfedpGGnmocL7pj7TSDA87qWpa4ytJWJY9rW0jir04Q4sm+9bllQJSTZh2/NhLKlLS1ta6yRZwZS9Aa39wlQ6VrDUtnOAgSkvUC8qSVKgjm0BO2y+8y/gr3oVrbd+Qe8tCCvNMHT89UdRsjeEfNagJElY1t9F1AN/hOEeBwmsez4U0vcQQseUtwUpNiUw0mSR9LE+/LWHMVfcjhyz8PupmUx1ihwOrvGI1t+Cr/IVPMd/jvv1b+F67u+Z+PmfMPG9zzLx/d/Bc/A/FxS0GZ4JPK9+Aykmgaj7/3DFxlCIgC3EKgerKXEUzttVcLF03eDVU+0UZcZRnBW6CeqB+rVOlCDTISeZcjdhf+hPQfOTdeY/+eLt0bT1jvNvvzyP16cBEGWyISGtmh02j+qhYaSZ9Ullizo/OSqRjUkV1AzVhXRdB9veJDM6nb2Zu0J63aU61zAAIOrXZuC78DKS3YGpKPQtfoWFMeVswnCPoQ+0Leg8rfUcUkzStKHkSloRxng/unts3vMNr3PqhsdfeWBhixaEFaZ11WA4h+ZsNrJUpswK7O/9v9gf/WvkmKRlex9h6ZS0YrTehkVnB3jPPAeyEtIaxalOkUEEbIau4Xrxn/Ae++nlh2aNoGsoSXmYK27HVLoPtfkU/trgdxE9R3+M4XESdefn5hxXEWoiYAuhAfcgPc7eZU2HPFPXT/+IZxl21y7XrwXRcORaSkoB9of/AskWQ/aF7/CHewwaOkf5+q8u4PNryJJMlMmGa5V0iawZbkAzNDYsYSxDQXweg55hRr1zjzwI1oB7iPbxTm7K2L6stZGLcba+n9SEKDKTQzskcrXTBtvQOi9i3nDXjG2xhZWl5GwEJNT280GfY6he1I6LgSfNV+2Qypfr2PQgBmir7RfA0FGy1gXmwY31L3jtgrBS/HWHwRI1taO8XGS7A+VyapsQuZT0Egz3KMb4wILP9Q91odYfwVxxB3J06GYST3WKDCIlUutrBK8T2x2/Tcwnv0PMB/8J+8NfIuquz2G7+YPYbvs4SmYF3mM/Dep3s7/lNGrDcSzbHkJZYkOehYqsO79VrmqgBmBJN/pzMQyDl0+2keKwsa00tOlnWncNmCzIKfmLOl+OS8X+8F8gJ+WSfelH/PG2EWpah/m3XwbSI6PNdpx+F7p7DLXjIr6Lr6F7QhPMhNrFgRpsio2i+PxFX6MgPvAPuWVs9tkzC3G+vwqALSmhr4NaCrdXpbplmG0lKRHTsTJS+C68AiYrloo7wr0UAZCj4pBTChZUx6Z1XgLNhylv27TXleR8kBS0vvnTItXWc0i2WGy3fRKQ8V0Mbaq0IISK4XOjNp/CXHiTaAIiAFzVZGnhaZHDh58C2YRly7tCvaygO0VqbedBkq976DZJkuTLv5vB8+Z350yZNzwTeN/+AXJSLpatoanHWwgRsIVQ1WA1afYUUu3LkxpW3zFKU9cY9+7KRZZDe3OsddWipJcGXb82E9kWi/3BP8aUt4XM5mf50oYWonrOc+TH38TmHGWs9QzOH30B94v/hPfIj1Br3w7hdxAahmFwcbCGisQSlHlm0c0lJyYTRVJoHl1Y+tVszvVXkh2TSXJUZKWPVDYNoukGW0tFOuTVdOcwasNxzOW3IlnFzmOkMOVuQu9rCvphkdp6FsxRKBnT06MlkwU5Offy0PLZGbqK2l6JkrsZOSYJU+FO/DVvhXS+kSCEitp8ClQf5tK94V6KECHkhGww2xbcFVcb6WKi6m3M6/eHvBYysK7gOkWqbRcCvRkss9f9y7HJ2G7+EFp3Lf6qV2c9znMkkAppu/1TS7pXXiwRsIWIR/VQP9y4bLtrAC8dbyUmyszejRkhva7W14Q+3BHU/LX5SCYrtrt/D/O6/aR0vcVvxbzJDvUMituHyxqFdfcHiHrgj8FiR1/EFvty65joYtQ3xvolprWaFTM5sVk0jS59h23EO0rTaCtbUjYu+Vqhdqaunzi7maLM0NVTrgWBX/o6lg33hHspwlVMOZsAA619/uH2hqGjtp7DlLNxxpTWwADtJgxdm/UaWk89+FxTHSYtG+8BvzuQdiYIEcZfdwQpPm0q5VcQJFm+3GRpYTtsvtPPIpksWDaHfncNgusUqU8Mog+1Y8rdPO/1TGX7UHI34z35FNoMQ7n9zadQG49j2f7ukM5TXQgRsIVIzXADqqEtW/1a54CT842D3Lk9G6t58Ts/V9P6GnG9/DVcT/8tWKMxF+4IyXUlWca696NEPfRn2B/9Mt13/T3N3mw6/SZGc2/DlLUOOTYJfWIwJO8XShcHA2mt6xIX13DkagXxubSNd6DNcUMXjPP9gdlRW1IjKx1S1XQqmwbZXJwc8h3f1czwufFVv4GpYAdynOicGUnklHwkW2ygrmweel8ThnsMU/7MtTxKWhGoPvSh2Qdoq63nQDZhujzSQUktRE4rxlf1Koa+sG6VgrCc9PF+tO4azCWhmb0mrB1KWjH6UEfQmQG6awS18QRx2+9dtqYcwXSKVC8/mFOCCNgkScJ268eRTFY8b3xn2oM43TOO9/APkZPzwjrgXQRsIVI1UE2UaWl1T3N55UQbFpPM/m1LL9JVe+pxvfjPuJ7+ClpvA5adjxPzwX9Cjk8PwUoDJEnClFGGkpJPRVEaWwszMWQfX/3xaToHnMgxyRjjkRewVQ3UkBubTbw1dsnXKojLw6/76ZzoXtJ1zvVXkWZPJSM6bclrCqWatmHcXo2tIa6nXO38tW+Dzx3yQdnC0kmSjJKzEbW9ct6ASW09F6h9yJl5RpSSerm2Y455bGrbOZTMciTzla7Blg33YIz1BWorBCECGIaB/9IbAJhDOHtNWBuU9BIwjKBnT6pt5wGDmA23LtuagukUqbWdD3T4dWQGdU3Z7sC67wn0/mZ8556fet17+EcY3vClQk6tL2zvvIbohk7VYDXrEsuWVPc0m+FxL8cu9nDLpgxi7YsvBFa7a3E9/w+4n/079IFWLLveR8wH/xnr1ofmzO8NhUyHA0x+dMPgH35yhnE5Fn1iIKIGyU74nLSMtbF+gcOyZ1N4ufFI0xIaj0z4nDSMNLE1wpqNAJytG8BqVliXF7ruT6udoWv4Kl9BSS9FSS0K93KEGZhyNoHXid7fNOdxautZlIyyWWsQpdhkpKi4WWs79JFujNHeqXTIqfcv2I4UnYivSrT4F8LL8HvxXXod11Nfwnf+RZTcLcixoh5ZmC7wt0wKuo5NazuPFJ24rPNB5+sUaWh+1M6LmHI3L2jH2Fy4C1PRTfhOP4s20IK/6R3UppNYtj2MctVol3AQAVsItI93Mu6bWLZ0yIOn2tENg3uW0Mpf7byE+7mvog93Yd39AaI/+M9Yt7xrxQb+2S8Pev79D1RgNSu8VuMGvwd8kdPq/9JQLQYGG5JDE7A5rPHEW+JoXkId24WBi+iGzuYIS4fUDYOz9f1sKEzEEqIU3bVA66rGmBjEvOHucC9FmIUpewNI0uWnwDPzD/egD3fO2dp8aoD2LE+d1dZzgffL3TL9PFnBsuEutK5qtMHQNCUShIXQx/rxHP8ZEz/5It7DPwTFjO32TxF11+fCvTQhAkmWKOTE7KAGaBuqb8ZRKMthrk6RWnctqD5MuTNnSMzFtvejSFGxeF7/9lWpkMtTi7cQImALgcqBaiQk1i1y0PJc3F6VQ+c62VGWSqpj8cGV78JLSFHxRH/gH7Fsug/JbA3hKucXfTlgi442+LOPbGPUiAGIqDq2i4M1xJijyY3NDur49r4JBkZmz+mWJImC+LwldYo8119Fki2BnJjImlfT2jPOyIRPDMu+htpyFhRLUEXOQnhIthiUtBJ8F1/Dd/G1GZuGOOveAbhud+xacloRxmjvjF0n1bZzyInZM+5YmMtvA5MFX+XsHckEIdS0/hbcr/w7zp/9Mf7KA5iyN2B/95ewP/plzKW3iFb+wqyU9BK0vsZ5U8kDgZJ3UYHSQs3VKVJtOw+KGSVz4Rspki0G260fRx/pwvC5wp4KOUkEbCFQNVhNYXweMebQt+9+81wXbq/GfTctfndNG+5Ea6/EvP7OFQ/UJtlNgWDT6XeTGGcjLScQFPlGImOIrG7oVA/WsT6pPKjB1IZh8G+/PM9/PV01Z1pnQXwug54hxnwLnznnVt3UDNWzJWVjxBWBv1PdhyxJbCoSAdskwzBQW89iyl4vbnwinO3Wj6Mk5eI98iNcT/0latuFaf+OXfXvICdkI8elznmdyTo2/ZpdNsPrROupv253bZJkjcZcegtqwzF019xtqQUhFHT3GK4X/ylQt771QaI/9C9E3fW5QMvzCPv7IkQeJa0Y/J45a8Yg8KAKxYKSuW7Z1zRXp0i17QJKZgWSaXH3vKbczVj3fBjb7Z8OeyrkJBGwLdGId5T28c5lSYdUNZ1XT7VTnuugIGPxnXb8lQdAMWNeF74BvpM7bE6/E4DyisCNTmdLZKQENY+24VRdQdev9Qy5GB730tIzTkPn7Ddck3Vsi9llqxyoRjO0iOsO6fVpvH2hi22lycREmcO9nIihD7VjOIdQ5tmVEcJPdmQQ9eCfYLvn9zB0DffL/4r7pX9BG+rA8EzgaauetTvk1ZSUApDk6+axqe0XwNDn3KGzbLgbdBV/9RtL/XYEYV7eYz8Fv4eoh/4M687HkaNF7bEQPCWtBJh7gLZhGKht51GyKlbkoeVsnSL10R6Msd4l7/JZNtyNuXj3kq4RSiJgW6LKgUsAyzJ/7cSlXobHvdx30+ILN3XPOP76o5hL9iDblt75cLEmAzaXGkghLCnJxm8oDHTN/bRmpVwcrEGWZCoSS4M6vrp1GACLSebVU7O39c6Jybo8QHvhdWzn+6uIt8SRHxeekztU1gAAYr1JREFUmR+zOXqxB6dH5e6dkfHUKVIEapakWXdVhMgiSRLm/O1Ev/fvsN78QbS+Jly/+kvcr/z75WBr/oBNMluRE3Ouq2NTW88hRcUhpxbOeq7syEDJ2YT/0usYmn/J348gzEbtuIjacAzLlgdQEoLrmCcIVws0WYqfs45NH+nCGB9Ysb+Bs3WKnKxPNuWsrdIEEbAt0Ts9Z0lfhpbrhmHw8ok2slKi2ViYuOjr+C+9AZof88bwDvC1T+2wBZqMmBQFrzke/+gAbq8azqUBV9Ja7ebg6gSrW4dJirOyf3s2Z2r7GRz1zHicWTGTHZtJ8wI7RXo1HxcHa9mcsiGoFM2VohsGB0+1k5ceS3GWGJZ9NbX1LHJqAbJdfC6riaSYsGy8l5gP/CPm9Xeh9TWhxCQip+QHdb6SVoTW1zRV22HoKmr7BZSczUjz/Nu1bLwXwz2G2nhiqd+GIMzIUH14Dv8AKT4Ny5YHw70cYZWSJAklrXjOTpFq6+VAaYVquGfrFKm2XUB2ZK65OaiRcyd4DVVXGfdNhHsZcxpwD9I42sKu9G0hzwGvbBqkc8DJfbtyF31tQ/Pjv3gQJWcjSkJ4m1ZM1rC5/Fe6QpodyTjkCc7UhbeObcQ7SudEd9C7pLphUNM6THleAnduC9TivX5m9l22wrg8WscWNkD70mAtft3Plghr53+peYjuQRf37MgRdQ9X0V0j6P3NYndtFZNsMdj2fJjo93+VjI/8zbzB1iQltWhabYfWUw8+97wNSwCUrHXICVn4Kg9E1IgTYe3wnX0OY6wP2y0fE7W1wpIo6SUY4/3orpEZv661nUNOykWOWfwmw0LJjumdIg2/B627FmUFmp6stIgN2H5Z9wx/evhv+fPD/5dvX/gBL7e8TvVQ3bQb/nA72XMGCYmd6fOnzizUyyfaSIi1ctO6xe/cqY0nMNxjWDbeG8KVLY4syUSZonCqV7oqRielkaw4OVHdG8aVwcWBGoCg69c6+iZwelQq8hJIirexrSyFN8914fXNHJAVxOcueID2uf5Kos12ih0FQZ+zEl491UF8tIWdFXM3Y7jRTKVgBFH3JEQ2OS4VS1LwaWNK2vQB2mrrOZBNmLLXz3uuJEmY19+FPtiG3t+8qPUKwmy0oU5851/EVLIXU9byN4EQ1rap33Uz7LIZngm03voV75AsJ07vFKl2XgJdXZOdmiM2YKsfaSYzOp2ShEK6nb081/Qy/3Huu/zR21/mb479Iy80hXbo6Pn+Kv636idB74IYhsGJnjOUOApJtIW2eLemdZiathHu3pGDSVncfyLDMPBVvoKckIWSNf+Nw0qINtunmo4AyLHJxEhu6poHGHP5wrauqsEaEqyOoNNaJ+vXynMD/93v2ZGDy6tytGrmgKxgsvHIWHCNR/y6StVADZuT1y/LIPbF6h50Utk0yB3bshb9c7lWqS1nkWKSkBOCGwkhrB1SXCqSLXbqJkZtO4eSWY5ktgV1vrl4N5gs+GveWs5lCjcYw9DxHv4BmG1Yd78/3MsR1gA5OQ8U04wBm9pRCYYRVGZBSNd0TadIre08mG1TTVLWkoi86/KoHvpc/WxL3cTH13+Iv775j/mnfV/m81s+xUOF9+GwOXix5SDn+ipD8n7jvgl+XP1LTved52zfhaDOaR5rY8A9yK6M7SFZw6S+YRf/9XQVqQlR3LZl8cXBWncN+mA75o33REzqWrTJjst/ZYdNjkkCIF6a4FRNX1jWpBs6dcONrEsqC/pzqm4dJi3RTmJc4IasKCuO/PRYDp7uQJ8hrSnB6iDeEht045HaoXo8mofNEZYOefBUByZF5vYtkTUTLtwM1YfWeWlFBoUKkUeSJOTUQvS+RvSRbozR3gXdtEiWKEyFO/E3Hsfwe5dvocINxV/7NlpPHbbdH0COWnyXaUGYJClmlJTCmQO21vNItljklJXNCrq6U6RhGKjtFzBlb0BSwj83LdQiMmBrH+/CwCAn9sqNod1spyKxlPvy9/P5zZ8kOyaTX9Q9PS0AWKxf1T+PV/ORYHVwoO1QULUEJ3pOY5bNbA3hTfWE28/XfhFIrfriezcTZV38D5zvwitItljMxTeHanlLZjdH4VSvpLRKlwO24gSNE5fCkxbZ5+rHo3mmdsHmo2o6te0jVORd2VWVJIm7d+bQPejiYvPQdedcGaAdXMB2rr8Km2KjLDFynhA5PX6OVHWze10acdGiDuJqWucl0HxBdRUU1iYlrRh9pBt/3RGABdcymstuBb8HtenkMqxOuNHorlG8x3+OklGGqfSWcC9HWEOUtGL0/hYM9UpWlKFrgUZLufM3Wgq1K50iuy6P1hlek+mQEKEBW9t4oIFDbtzM6UWKrPDh8vcw5pvg6cYXlvReNUP1vNN7hnvybueBwnvonOjm0lDdnOf4dZUzvefZnLIemym4tJf5+FWNb/zqAoNjXn7v8Y2kJdoXfS19pAet7Rzmdfsjqsg42myfVoMoxwSGLm/JkKjvGJ210+Jyah0L/KzlxQaXytbaM47Xp00L2AB2lqcSH2Ph1XfaZzyvID6PAc/QvI10NF3jwsBFNiZXYJYj5wnRW+e78Pl17tohUv6upbaeDaRgZARXAymsPUpqEQC+qleRE3OQYxc2UF5JL0WKT8df+/ZyLE+4wXiP/xRUH9Z9HxO7/kJIyWnFoKtoA1ceQGu9DeBzhSVQutIpshO1LZAhp+RsXPF1rISIDdgc1njiLLPPDcuNy2Z/7j6OdJ2kbrhx1uPm4tf8/Lz2N6REJXFv3n52pm3BYY3n1da5B5leHKjGpbq5KT006ZC6YfA/L1RT3zHKpx9aR0m2Y0nX81UdANmEed3+kKwvVOzXpERKMQmARJEj0Nb/ZM3K77K1jrdjVSykRwfXRGOyfq0s1zHtdZMis39rFlXNQ3QNOK87ryAusIPXNM8uW/1IE06/K6K6Q2q6zuunOyjPdZCbFr5ZfpHIMHTUtvOYcjauyRQMITiBAdoSqN5F1XBIkoS57Fa0njq0ka75TxCEWagdVagNxwMz1xxi5poQWpONR/SrBmgHGi0pmLLDc98y2SlSazuPnJyPbHeEZR3LLWIDtmB2PB4suIdkWyI/rfkVvkUMHn2l9Q363AN8oOwxzIoZk2xif84+6keaaJmjQcSJnjPEW2IpSyhe8HvO5DdvNXGyuo/33lHEzvKldd/T3OP46w5jKt4dcfOgos12XKob3QjMK5JkE1K0gyh1lIKMuLCkRbaMtZMbmx30rLPq1mGyU2KIs1+/c3nb1kAzjoOnr2/xnxMbGKA9188VwJm+C1hkM+uSyoL7BlbA2boBBse83L1DDMq+lj7QiuEaEe38b3CSJWqq4cxinzKbS/eApOCvEbtswuIEZq79ECk+HcuWB8K9HGENkqPikOLTpg3Q1trOo2SUI1mCm2Mb8jVd7hQZ6FK59tr5T4q4gM2teuhzDZATRMBmUSx8sPxx+twDvNRycEHv0+Ps40DrG+xM20r5VbVCezN3YTdF8WrroRnPm/A5qRqsZkfa1pB08HvzXCcvHGvl9q1Z3Lcrd8nXGz97EFRfRLTyv1a02Y6BgVu9kvooxSRhjA9y07o02non6B68fndquai6Sud416ypt9fyqxoNnaPXpUNOirNb2L0+jaNV3Uy4pz9AsChmsmMy56xjqx9u4mjXSXakbcWiRE4q64FT7aQ4bGwuXlia141AbT0LkrRmc+aF4CnZ65Fik5FTCxd1vmx3YMrbjFp/BENXQ7w64UbgO/fC5ZlrT0RUOYSwtkwO0DYMA32sD32kK6yBknzVTvJa/lsccQFb+3hg+GiwN9HliSXsztjBwbY36RgPLpXEMAx+VvtrLIqFx0oenPY1m8nGrdl7ON9/kV7n9Z0LT/edRzd0bgpBd8jKpkF+9EodGwuT+PDdJUvONTd0ldFTL6JkrUNJirzdkMnh2c5r6tj0iUF2lqciwYrusnVOdKMaGvlxwQXKjZ1j+FV91oAN4O4dOfj8Om+fv/5nsSA+l9ax9hlHR4z7JvjexSdJiUri8Wt+JsOpuXuMho5R7tyegyzfOLUQ+kgPzmf+74zdsK6mtp5DSStBssWs0MqESGXd9R6iH//KkoruzWW3YrjHUFvPh3Blwo1AH+3Fd/4FTEW7xcw1YVkpaSUYnnGMsb4rM0jDmGUiJwYaFEq2WOTkyJpdG0oRF7BNNRyJDb51+GPFDxJtsvOTmqeCmqN2ouc09SNNPFJ0/4x1crdn78UkKxxse3PGc7NiMsiKyQh6fTPp6J/gv56uIjslms8+vB5FXvp/Cq29Cm18CMuGu5d8reUQbQ40UnFd1SlSjk3CcA7hiDZTluvgxKXeoLp0hsJCG45Utw4jSVCa45j1mJzUGMpzHbx2pgNN16d9rSA+D5/up9M5fV6bbuj84NLPcKouPrHhIyFrZBMKB0+1Y7Mo7Nu0tJ/31cZ34SX03gZcL38NbXjmB0H6xCD6YNuKz50RIpMkm5acEqTkbESyO/DXiplsQvAMw8Bz9Mcgm7De/IFwL0dY4yZnnGm99aht55Hj05Hjg5tjuxykqHikqHiU3C1IIbiXjlQR9521jXWQYHUQawn+iXW02c57Sx+mbbyDNzoOz3nshM/JrxuepzA+jz2Zu2Y8JtYSw80ZuzjRc4YR7+jU673OPlrH2tmVvi3otc3E41P55tNV2CwKv7/E9v1XUzsvIpksKGEq/JyP/XLA5pzWeCQJdA3DPcpN69LoHXbT2ju+IutpHW8nxhwd9ODz6rZh8tPjsNvm/u91984chsa8nKkbmPb6ZOOR5tHpdWwHW9+keqiO95Q8RE5s5BSJj0x4OVndxy0bM0L2M7rS1O5aXM/+P3TncNDnGF4n/vpjgRbFsgn3i/+MPnH9uAa19RwAigjYhBCRZAVz6S1o7RcW9DMr3NjUltNo7ZVYdzy2ZhsuCJFDTsgAix21vQqtqybsfwMlScL+8Jew7flgWNex3CIuYGsf7ww6HfJq21I3sTG5guebDjDgHpz1uN80voBb9fDBssfnbDRxZ+6tGBi83n6lAPxkzxkkJHamLW3e0o8P1NEz5OIzD60nIda6pGtdTeu8iC23Akkxh+yaoRQ9FbBdqVObHJ5tjA+wvSwVRZZWLC2ydaydvLicoFJRPT6V5q6xOdMhJ20uSiY1IYpfv9WE23ulFiXR5iDOEjstYGsYaea55lfYlrqJWzJ3L+4bWSbPHmlB1w3uXKWt/A2/B8+h76D11OE7/2LQ5/lr3wbNh3XHY0Td/4cYPhful/4Fwzu9vlJtO4cUn4Ycf2PtPgrLy1x+KxiGaPEvBMXwe/AefRI5KQfz+jvDvRzhBiBJMkpaMWrTCdDViKgbk+NSkSyLH4e1GkRUwOZW3fS5B8gNMkXtapIk8f7SR1EkmW+c+y7fuvA9vlv5I7538cn/v737jo+rvBI+/rtliro06l2ucu8G00wHU4yBhEBCgCwppG6SNwlxdjcJyb55QwiQbMIum0IKBBICJPTebTDGvXdbsnrvoym3vH+MJCyrSyNpJJ3v57OfjWfu3HnmYTQz5z7nOYdH9v+dvx58iscOPskHFVu5OHc1WbEZ/Z4vJcrDsrRFvFe2GW8wVNlwc+V25npmk+CKH+5LZOPuCt7fW8k150xjziB+/A+W5W3EaignqiByK+TE6B0pkd1W2EKFLKzWOmKjHCycnsyHB6qxRjkt0mf4qGyrHnQ65JHSJkzLHlTApqoK/3LFHKobvDz8yqGuFM+uBtrNocIjrYE2/rjvMTzuJD415+MR1S/njW2lvL2jjEtX5pKeNDE/BP0fPoHdUoeaNp3ggXew2psHfIxtWwT2v4mWPgstJR8tJZ+oy7+O1VRF+8u/7GoWagfaMcsOoOctiaj/bmLiU+PT0LLmEjy0Adu2Bn6AmNL8257BbqvHfc6tKGEohCbEYGjpM8G2wRmFljFr4AeIEYuogK2r4MgQ9q+dKsmdyC3zbiTWEUODr4kKbzVFzSUcajjKrtp97Kjew4yEAq6cdsmgzndp3gX4TD8byjZxtPEEDf7GEaVDltW28ZfXDjEnL5G1ZxcM+zy9Mcv2AxA1LXIDtqiOvVltp+1hg1DABnDGvDQaWvwcKWkc1bGUtJRhY5MfP7jiLAeKG9BUhZk5g2uVUJiXxLXnTWfz/ireOaUAybT4PGrb62gOtPDwgcdpDbTy2QU3d81NJNh5pJbHXj/M0lkpfOLC8LSuGGtG+UGC+97AseAS3Od/DswgwT2vDvg4s2QvdnN1tyvVetZc3BfdgVl1FN8bD2JbJkbZvtCVRUmHFKPAUXgedksNZsWh8R6KiGBmfRnBPa/iKFwtP5rFmOrsx6bnLERRJ+aWiYkmoma5swjEcFbYOi1JXRC2psM5cVnM8xTyVslGytsqcWlOFqfOH9a5/EGT/316L26HxheumR/2intG2T4UVyzO9ALopXFzJNBUjSjdjfeUKpGKww2uGOyOgG1JR+n4QycbKcwL3wrk6Yo7itsMJWCbkZ2AyzH4K5hXnZXPkZJGHnvtCNMz48lLj2NaQmgf25/3/Y2DDUf4xOxrR/R+D7eiymb+99m95KfH8YW14X+fjgXb8ON79w8ocam4Vn4cxeFCn7acwL43cC65st+0icC+11GiEtCnreh2u2P6SuxzPo3/vUfwb/wztmWCK0Z+JIlRoU9bAe89QvDgO+hZc8d7OCIC2baN/72HwenGeeYN4z0cMcVo6TNQUwpCKdxiTETcCpvHnUSsM2a8h9Ll0vwLaAm2srVqJ0tTFw27P9ZfXz9MeW0bn187n8TY8O1bg9AHt1l2AC177ohKSo+FaD26W1l/CO1js1pCBTrcTp24aAeNrf5RHUdxcwnJ7qRBFbdp9QY4WdkyqHTIU6mKwufWziMu2sH/PL2Xdr/R1aT7YMMRlqYuZHX2WcN9CWFX29TOfz2xm7goJ1//+CJczomZXuP/8KlQL6Lzb0dxhP7WnEuvhmA7gX1v9vk4q7kas2QPjrkXoGg9r2U551+Mc+laggffxTjyPnquXFkUo0PRnThmno1xYmuPvZNCALTufQez4hCuM25Adfesdi3EaFJ0FzHX34UeoUXuJqOI+nVf3FIaUasNALMSp3f16Tozc3jpkB/sq+TdXRVceVY+86d5wjk8AOymSuy2erTs4a3+jaUYR3S3lEgIBWydK2wAibEuGlpGP2DLG+Tq2p5jddgw5IANQs2077hmPrWNPv700kEcqk5BfB4pbg83z42cfWteX5D/emI3AcPiG59YTEKYLyqMFaPyCMG9r+GYd1G3lQktpQAtdyHBPa9gG72/twL73wRFxTH3gj7P71xxfVdRCL1gZNViheiPY85qMA2CRzeN91BEhLH9bdS/8TBq2nRZ4RBiioiYgM0b9FLbXjfs/WujRVEUPjZrLedmr2Jm4vQhP76y3sufXznErJwErj1vdBr6GWX7ACZEs8wYR3S3oiMASlyoeXZncY7EWBeNrYFRG0NLoJU6X8OgC47sPlqDU1eZnjW8YjOzcxO5bvU0thys5q0dZXxh4a18Z8XXiNJH1rMpXAzT4r//uZfKei9fvW4B2SmRs8I9FLYRwPfOQyixHlxnfqLH/c6la7F9LQQP9uxxZRt+ggffRZ+2HDWm78BcURRc595G1NrvoU9bGdbxC3EqLSUfNTmf4IF3Qim4QgC2ZeLb9BimtwX3ubdFfFaNECI8IuYv/WRnwZFhlPQfbdMT8vlk4fX9tgHoTdAwefDpvTg0lTuuCU9z7N6YZQdQ4lJQ4lJH5fzhFK1HddvDBh2l/YM+CIRuT4x1jmpKZGdz9oJBrrDtPlrLrNxEdG34//2uWJXPohnJ/O2NI9TVWxGT9mvbNg+/fIgDxQ185oo5zC0I/wrwWPFv/Sd2UyXu8z8b2ht5Gj1jNlrGbAK7XsI2jW73BY9+AAHvoMpiK6qGnlkYMaujYvJyLrocq74E31u/k6BNYNaX4X3m/2Icfo/Es9ahpeSP95CEEGMkYgK2zgqRuRG2wjYS7++tpKS6lduvmosnfnSqANqWhVG+Hz1r3oT4AdlbSqTS0Yutcx9bUpyL5rYApjU6Ja2LmktQUAb1XmtqG97+tdOpisJnr5pLXLSTB5/ei9dnDPygMfDm9jI27qngmnMKOGfhxO0nZlYfI7jnZRxzLuh3pdm59GrstnqCR97rus22bYL73kD15KJlzB6L4QoxKI5ZZ+M84waMYx/ge+u3ErRNUbZl4t/xPN5//BC7pRb3xV/Gc+Gnx3tYQogxFDEBW3FLKcluD7GOgVceGlv9HC8fuKfSeCupbsXt1Fg8I3nUnsOqLYJAO9oESIcEiO5IibRO6S/U2Ty7s7R/YqwLG2g6LS3Ssi1+svl+frn9f9lVs6/bOYbiZHMJGTFpuAdRSv9gcQMwvP1rp4uLdvKldQuob/bzxNtHR3y+cNiwq5zpWfGsO3d00nXHgm0E8L39EEp0Eq5VN/Z7rJazEDUln8DOF7E7LgiYVUex6k7imH/xhLjoIaYW15KrcJ7xCYxjmyVom4LMhtCqWmDLk+j5S4i+4Sc4Zpwx3sMSQoyxiAnYSppLB71/7S+vHuYnD29lwyn9rSJRRZ2XzOSYUf0RaHT0X5soAVuMIxobG5/h67pNiQuV8rdPCdiAHvvYqr21lLdVUtxcwm/3/JkfffBz3irZ2O1cA7Ftm6LmEvLjBl/OP8atk5c+cDXJwZiZk8D8aR5ORMAFh4YWPyerW1k2O3VCByqBnS9gNZbjXv0ZFGf/+wIVRcG55Grs5iqME1sACO57HZxROGZGTsVOIU7lWnIlrjM7grY3fyNBWwSxbYvg8S1Y3sbwntcy8e98Ae9TP8RursF98ZeJuvSrqFHD20sthJjYIqImdbvRTq2vnnOyzxzw2KBhse9EPbqu8seXDmIDqxdnjf4gh6Giro35o7wnyCzbh+rJnTAf4jF6qAdWW7CdaEfofyvuONAcXStsSXGdAVv3fWylraEA/ZvLvkStr563Sjbw5JFnef74q5ydtZILcs4hOar/+a73NdIabCN/EHslbdtmf1E986enhHX/YXK8OyJWiHcdC6WgjuYK8GizmioJ7HoBfeYq9NzBNY3Xpy1HTcwksON5tIzZGMe34lhwSVcLACEikXPxlYCCf/Pj+AD3RXegqBOz9cZkYQe8+N76HUbxDvTpZxB1yZfDdu72V3+FeXIX+rQVuM69dcJ8xwshRkdEBGzlrVXA4BpmHy5txB80+fK1C3h3dzl/eukglmVzwdLI2vvm9Rk0tgbISO67Se9I2UYAs+oIjnkDF0qIFNGO0AqI1/ACoUBBURSU2GTsjj1sibGhXnenB2xlrRVoikZWbAZ58TksS1tEUfNJ3irZyNul7/FWyUZuLLyO87JX9fn8xS0lwOAaZpdUt1Lb5OOmy9J7vd82g6AoQ+7FlRTnorU9iD9oDqkRd7jtPlpHcrybrIlaFdK28b33F1AduFbdNOjHKYqKc8nV+N7+Hb43HgTbxDnvolEcqRDh4Vx8BYAEbRHArC+j/bVfYzdXoybnYxRtx/a1orhHno1hNpZjntyFc9k1OJdfN6EzIIQQ4RERKZHlrRUAg0qJ3HOsDl1TWTg9ma9dv5BFM5J5+JVDvLm9dLSHOSQV9aFmp1nJo/dj2Kw8AqYxIcr5d4pxdK6w9dI8u2OFLS7aiaooPXqxlbaWkxGThn5KgFQQn8e/zP8UPz5rPbMSp/P00RdpCbT2+fzFzSXoikZ27MAFNrYfrkEBzpzf+7HeZ36C77X/7mpHMFie+I4VxFHuNdefoGGyv7iexTOTJ+yPAePEVszSvbhWXo8anTikx+ozz0SJS8GsPIyWuxA1ofegXIhI41x8Ba5VN2Ic/zC0p22Inz9i5ILHNuN9+kcQ8BJ19XdxX/BZsMLXM884sgkUBce8iybs57MQIrwiI2BrqyTF7elKkevP7mN1zMlLxOXUcOgaX7luIUtmpvCXVw/zxrbICdoq60IBSeYorl6YZftA0dAyC0ftOcItWu8jYIv7qHm2qiok9FLav6ylnJzY3tNfk9yJ3FR4HQErwHPHX+7z+YubS8iOy+oW9PVl2+EaZuUmkhjXM1XObCjHqi3CKN6BWbxzwHOdyhMXKnZS3zz4vXfhdvBkI4GgxaIZKeM2hpGwA+34Nz2GmpyHYxirY4qqd6SYgXP+JeEenhCjyrnoCpxLrsY4thm7pWa8hzNl2JaBb9Nf8b3xIFpyPtHX/wg9sxAtOQ81pYDgoXdHHEDbtk3w6Ado2fOHfCFKCDF5RUTAVtZaMaj+a9UNXirrvSw8Zc+NQ1f58nULWDorhUdfO8xrW0pGc6iDVl7XhqYqpCaOTjl/AKP8AFr6jF57TkWqrhW2HqX9U7Dbm7GNUKGR05tntwRaaQq09Lsy9vbmJjKtebxX/iElzWU97rdsi5KWskEVHKms91JW08by2b33tjNObA2NOy4V36bHusY9GEkdK2z147jCtutoLU6Hytz8xDF/7nAUTPBvfwa7rQH3ubcOOyXMMfcCoq/9AXre4hGPR4ixpndUCjSrj43zSKYGy9tI+ws/J7jnFRwLLiXq6u+ixnxUPdhReB5WXQlWbfHInqfqKHZLjRRBEkJ0ExEBW6O/aVD71/Ycrwdg0WlFEnRN5UvXLmD57FT++sYRXts6/kFbRa2XdE/0qDXLtn2tWDVFE6Y6ZKdovWMPW2/NswG7NfTf+PTm2WUdabN9rbD5AgavfFjC0e3p2EEHP3/3Uf757jEq6tq6jqny1uAz/YNqmL39cOiq9fLCPgK2om2o6TNxr/4X7JYaArtfGvCcnTxx4xuw2bbNrqN1zMv34NDHdv9L8OC7FP/X57Dah190xawvIbjnVRxzzkdLnzns8yiKipY2fdiPF2I8qUnZoLswqyRgCzfb34ZRcYjAvjfwbXwY77P/j7bH12NWn8B94Rdwn30zitY9S8MxcxVoDoKH3h3RcwePbgLNiV6wbETnEUJMLhFRdAQGV3Bk97E60pOiSE/qmTqpayp3rJvPr57czVPvHOPi5Tmo45j7XVHvJSd19NIhjfIDgI2WPX/UnmM0aKqGW3PjDbZ3u72reXZrLWpiBolxLg6XNHbd31khMjuu9xW2uqZQeuHNF82n2LDZ1vYmLx7YzHPvZ5CfHseZ89KJzqoEGFSFyG2HapiWGddrw3OruQarthjXqhvRs+ehTz+DwI7nccw6GzWu9wDvVA5dIy7aQcM4pUSW17ZR1+zjqrPzx/y5gye2YHmbCex8AfdZnxzy423bwr/hYRRnNK4zbhiFEQoxMSiqhpY6bUqusFm+FtpfvBfXmTeGbQ+3bQTwvftHzIqD2G0NH93hjELz5OKYeRaO+RejeXr//lBcMejTVhA8ugnXqptQdOfQx2AZGMe3oOcvGbBFiRBiaomYgC13gIIjgaDJwZMNnL8kC6N0H8Ej76NExaFGxaNEJaBEJ6BExXPGNDf7Thg0NPtJThifVEHDtKhpaGflnIF/vA+XWX4AHG60tInX8DjGEdUjJVKN69k8u81nEAiaOB0apS0VJLoS+mysXtsRsOVnxHFB5qVUbNmPd34R5zpWsfVgHX9/6yjZS07gjnKRFt3/f5f6Zh8nKpr52Pm9r74YRaF0SL1gBUCoAMDJnfg3/Y2oy742qDlIinON2wrbrmOhOV40fWzL+dtmELP8EKg6wf1v4ly0pltK0WAYh9/DrDqCe/XtYanGJsREpqXPILDrZWwjMKwAYaIyT+7Gqi3G99Zvifn4/w3LZ4F/8+MYRzehz1yF6slD8+SgerJRYjyDLvzhmLMa4+gmjBNbccw6e8hjMEv3YvtacMySdEghRHcREbB53Eld5d77cvBkI0HDYtGMZPxb/hurvgRQwOy+d2gx8K14D5X1y8YtYKuq92LZNpmjWCHSKNuHllk45JLykSDaEd2j6IgSkwSK0lV4JKmzeXZbgLTEKMpay8npZ/9aZ8CWmuBGUzU+PusafrXzt6g5RXz/zIt44q2jvNmyiemeTFSl/zTVj9Ih03q9P3hiG2pyPmp8KPBTY5NxLl1LYMtTGKV70XMWDDgHnjh315jH2u6jteSlxfa6ejiazKqjYAZIvvRfqHv9zwR2Po/7nFsG/Xjb14p/899R02eiF547iiMVYmJQ02aAbWLVFqNlzBrv4YwZo3QvONzYvhZ8G/+M++Ivj6iaonFyJ8F9b+BYcBnusz817PNomYUocakED20YVsAWPPIBuGLQchYOewxCiMkpIvawZQ2ixPqeY3U4HSqzk0ysmuM4l19L7O2/IfYzDxJz491EXfNvuC/5CnbuUnL1eqpqm8Zg5L2r6KgQOVol/a3WOuymqglVzv9UMXp0jz1siqqjRCdhtXSssMV19GJr8RO0DCq91WT3sX8NoLapHYeuEh8TelyhZyaLU+bzcvGbNPmbuXB5Jkp0M/6muAHHt/1wDdkpMWR4eqbeWm0NWFVH0act73a7c9EalPh0/O8/im0aAz5HUryLhpaxD9ha24McKWti0cyxb5Ztlu4DRSVu8UU4Cs8leOCdrhXVwfBveQrb34b73NtQBgi6hZgKOvdgTqW0SNu2MMv2oectwbn8OozjWzBGUE7f8jbie/shVE8urjM+PqKxKYqKo/A8zPIDWM3VQ3qsHfRhFG/HMf2MHvvjhBBiUL96nnvuOa688kouu+wyHn300R73HzhwgOuvv57LL7+cf//3f8cwBv7Beqrs2Ix+77dtm93Ha5mX74GSHQA4pq0INVx2RqEmZKBnzMYxfSXuGaE0tZbqiiGNIZw6C1309oM/HMyy/QATruBIpxhHdI+USAitVNmtnc2zO1bYWv1UtlVh2RY5cf0FbD6S493drrJeN/NqLMvk2WMv00Y9impTUqTT1NZ3Rcdmb4BDJY0s66s6ZNF2APRpK7rdrmgO3Gd/CquxguDe1/o8fydPXCjl0x8YecXEodh7og7bhsXjUM7fKNuHljYD1RWNc+k1gE1g+3ODe2zlEYIH3sax4FK05IGLxggxFajRiaF+glMoYLPqS7Hbm9Fz5uNcfCVq+kx87z0ypIs/nWzbxvfOQ9hBH+6LvhiWtFLH7HNBUQge2jCkxxlF28EIoEs6pBCiFwMGbFVVVfziF7/gscce4+mnn+bxxx/n6NGj3Y75zne+ww9+8ANeeeUVbNvm73//+5AGkRXTf8BWWe+lptHHwhnJGMe3oibn9tnoVo0PpbH5GoZ2dSucKuq8JMe7cDlHpwKfUbYPJSoeNWng4hmRKNoR3aPoCIASl4zVVSXyo+bSpR0VIvsr6V/b5CPltBTY1OhkLsw9jw8qt/JuWegKrNESz+v9VBHdeaQW2+6nOuSJraiJWWhJPYNHPW8xWt4S/NufwTp103ovOtMR68d4lW330TpioxxMy4wf0+ftqmqaEyqSo8al4JhzPsFDGwa8Em211uF77dcocSm4ll87BqMVYuLQ0mZMqUqRZuleALScBSiqStSFXwDbxvf277Fta0jnCu57HbNkD65VN6J5+t9HP1hqrActZyHBwxuxrcGPJ3h0E0ps8ogq3wohJq8BA7b333+fVatWkZiYSHR0NJdffjkvv/xRY+KysjJ8Ph9LliwB4Prrr+92/2BkxvYefHXa01EkYWGmill1tMfqxqnUuNDKgd1SO6QxhFN5Xduo7V+zbRuzbD9a9rwR5eyPpxg9iragF+u0L1c1NgW7tR7bsohx6zh0lcbWAGUt5ThVB6lRfafx1Ta2k5LYcx/k5QUXEeeI5YOKrcQ5Ylk6LZe3tpfR7u99FXj74RpSEtzkpvXcxG75WjArDvVIhzyV++xPgWXg39z/RYvxKO1vWhZ7jtexaEYyqjq2753eqpo6l64FVcG//dk+H2cHfbS/8ktsI0jUmm9I5TQhTqOlzcBuqx/wItFkYZTuQ03K7ipYpMan4Trrk5jlBwjuGTi7oZNZX4J/8+NoeYtxzLs4rGN0FJ6H3daAWbpnUMdb7c2YpftwzDxL0r2FEL0a8JOhurqa1NSPVhvS0tKoqqrq8/7U1NRu9w+GHeh/GLuP15GVEkN83T7ARp+2ss9jlegELEXDHWggEBzbdDMAy7aprPeOWsBmNZSF0kGyJmY6JIRW2GxsfEb3YEWJTQbbxPY2oigKibFOGlr9lLaWkx3bd7GQdr9Bm8/oscIGEKW7WTvjcgDy43O5alUBXr/BOzvLexzr9RnsL6pneWFqr8GwWbQDbKv/CwbxaTgXXRGqFFZxqM/jkjpW2Bqaxy5gO1bWTJvP6NHHcCyYZft6VDVVY5JwzL0I48h7WI2VPR5j2xa+N3+DVV9K1CVfQksKzxVwISYTLX0GMDX2sdmGH7PyENpphZ0chavR85fi3/IEZn3pIM4TwPfGb1Cc0bjP/2zYL37q+UtR3HEEDw6uJ5txbHPou0WaZQsh+jDgzlbLsrp9mNm23e3fA90/GFs3vse1N63t9b52v8HhkkbWnjcDpXQDjpQcMmYX9nu+higPHl8rQUUlO3XgIhPhVF3vJRC0mFXgIXUUnrv5ZDFeIHXBchxJvZ9/NJ43nDJaPQC44xVSYz8aqzc7h0ogQW/HnRpHalI0rf4gVW0VnJW3os/XVVQRasI8PSep12OuSb6IkvYSVmYv5ozcbBbNLOKN7aXctGYuDv2jIPCd7aUYps3FZxT0OE9qahwVb+xET0gjfc78ft/j1qWfpPTYJgJv/5aUW36MI6lnym9iRy9Bv2WP2X+vFz8sQVMVzl+ZT2yUY0yes9PJigNEFywkLT10VbzzNRsX30jJoXdQ9r1A6rXf6PaYujcfwSjeQfJlnyVh2TljOt6JKNL/7ieaiTKfdtJ8Tmg6rpYSklMvHO/h9Ckc8+k9dpRW0yB5/kqiTzufed3XKPntNzDe/T3p/3I3it73Z1ztKw9hNZSScdN/EJ03OheC6hZfQNOWF/FEW2gxCf0eW1b0Ic60AjIK5wzpOSbKe3SikPkMP5nT8BkwYMvIyGDr1q1d/66pqSEtLa3b/TU1NV3/rq2t7Xb/YCRVfsgL787ljLk9UyN3HK7BMG1me8C3Zx/OpWupqWnp/4QxyXiaazhwtIYYfWxTv/YeD6VvxjnVgcc5DO3H9qNExdMQjEbp5fypqXGj8rzhZPlCQVJJVQ1q+0erYqYVCmLqS0twuLOJcekU1VXRltxOspbS5+s6ciI05w7V7vOYm2aEqn/V1LRwybJs7v/7Lp575wjnLfpoL9rbW0+SEOPEE6N3O09qahzVZVW0n9iNY8El1Na2DvganZf+K97nf0bpwz8g+ur1XS0AThUf7aCkonnM/ntt2lPOrJwE2lt9tLeO3d45q7kao7EKbd6l1NS0nPYe1XDMu5jWXS9hzV3TtY8keGgDvk1P45h7If78cyP+PT3eJsLf/UQy0eZTTc6ntegAVoSOOVzz6du3BVSd1uhc2nqcT8V17r/Q/up/UfbKI7jOuKHXcxgnd9G+9UUcCy6lLX5mL+cJDzN3FWx+jsoPXsW5aE2fx1lNlfjLj+A68xNDmqOJ9h6NdDKf4SdzOjSqqpCc3HdPyQFTIs8++2w2bdpEfX097e3tvPrqq6xevbrr/uzsbFwuF9u2bQPgmWee6Xb/YEzTa3jt1U009LKfZ/fxOtxOjfzgUbD7T4fs5EpKJ1ltpaK+ZyXC0dZZ0j8zZXRSIs3qY2hpMybs/jWAaD0UmJ1eeESNDe0/tDoqRSbFuWi2QsFYfxUia5pC50mt343V1DO17nTzp3nIS4vl5c0nsWwbCDVm3328jmWzU1F7mVvj5C6wDBz9pEOeSkvOI/qqO7GDPrzP343VUtPjmKR4d6/v+dFQ29ROWU0bi8alOmRHVdOc3tN4HYuvAIeLwLZ/ho6vOIRvw5/QsufhOufmCf1eF2IsaGkzMGuKsK2hVWieaMyyfWiZs1F0V6/36wVLccxZTWDni7Q99UPanvw+bU/8O22Pr6f1b9+l9a/fof21/0b15PQZ0IWL5slGTZtO8NC72B3fM70JHv0AUNBnrBrV8QghJrYBA7b09HS++c1vcuutt3Lttddy9dVXs2jRIj7/+c+zZ09oQ+29997LT3/6U9asWYPX6+XWW28d0iBszcEqbS9/fPFAtw8227bZfayO+QUerOLtKPHpqJ6BKyPqianEqn5qaxqHNI5wqKhrI8atEzcKKWe2rxW7qRK1Y8/CRBXT0SS9LdjW7XbF4UJxxWJ39mKLdWG6GlFQ+q0kWtfkI0Y3Ud7/A+1v/rbfL0cARVFYsyqPijovu46EgsN9J+oJBC2W9VUd8vhWlOhE1I6+R4OhpeQTfdV3sAPteJ//GdZphXA8ca4xqxK5u6Nwz+Jx6b+2FyXGg5rQe5VP1R2Hc8GlGCe2EjyxDd+rv0aJSyXqkq9MyMbwQow1LX0GmAGsQezfmqgsbyNWfSla9oJ+j3Ot+iSOOeehxCSixqWgJmSgJuehpU5DS5+JY9bZRF36tbCU8B+Io3A1VkM5ZuXhXu+3bZvg0U1omYWosZ5RH48QYuIa1K+htWvXsnZt9z1mv/vd77r+95w5c3jyySeHPQjH9JWsaHieZ4rKeGtHChctCwVlZTVtNLT4WbIqCnP7AZyL1wzqarsaF/rR7WsYWvGTcKiobSMzJWZUVgXM6uNA6GrqRBbjCK0+Ngaae9ynxCZ39dNJjHWiRreQ5EzC3ccVVQiV9C+M9wI2Vs1xjBNbcEw/o98xrJyTxj/eOc6Lm4tZMiuFrYdqiHHrFOYm9jjWCvoxSvbgKDxvyBW8tJQCoq+6E+8L9+B9/mdEr12PGhsKmjxxbg6ebBzS+YZr97E60hKjRq03YF9sy8IoP4Cev6zfvwnnojUE9r2O77VfgyuGmDXfQHGNziq1EJNN53eCWXUMLaVgfAczSszSfQDoOfP7PU5xRuFefftYDGlAjhln4t/8OO3P3Y2aPgM9fyl6/lLUxEwURcGqOYHdVIW++MrxHqoQIsJFRP1Yx+xzUG2T6zJK+PubR7saT+/u2A82Xy8B2xxUOiR8VNrfbK4ZcLUl3MrrvGQlj1LD7OpjoChoqdMGPjiCxTpiyIrJYFvVrh73qXHJ2K0frbAp0S0kOXpf9epU29TO9OjQvjIlOhH/h09im/2nBmmqyuVn5HGsrJmDxQ3sOlrLkpkp6FrPP4n2YzvADPRbzr/f50otIPrKb2P7W/E+d3dXQOqJd9HuN/psMRAu/qDJgeIGFs1MHvP0Qqu2CPxtA//IcsXgXLIWVJ2oS7+KmtB/b0YhxEeU2GSUqIRJXSnSKN0b6j+anDveQxk0xRlFzHV34Vx+LZgGgQ+fwPvEv9H2+Hp8H/yNwK4XQdUHnWovhJi6IiJgU+PT0XIXsVzZj1u3+f3z+zFMi93H6shLi8VZsRMlLgU1JX9Q51M6VthizCZavMHRHHo3Ld4Are3BUSvpb1YfQ/XkoDh6lq+fSBRF4eysMzjZUkppS/fy+p0rbLZtEx0NqttLDP2n8dU1+cjSGkB34V79GezmaoIH3hpwHOcuyiQ2ysHvXziA12/0mQ7ZdmgziisWLbP/6qT90dKmE33ld7B9raH0yNZ6kjp6sY32PrYDxQ0EDYvF47J/LXRVXMseuA2Fc/EVxN76K/SsuaM9LCEmFUVR0NJnTNqAzbat0P617PkTrk+ZmpCOa/k6Yq6/i5hP3Y/r3FtRE9II7n0d48RW9LzFkk0ghBhQxHzyORdehuJr5stLfZyoaOHvbx3laGkTSwtiMEv3ok9bMejVAcUdh6U6SVbbqBzDwiNdBUdGYYXNtq2ugiOTwcqMpeiKxvsVW7rdrsamQNAH/jbalHoAHMHEPs/j9YV6sCXbdaiebLTcxWhZcwlsfxY70N7n4wBcDo1LVuTQ0OLH5dCYX9BzD4FtBmk7shW9YBmKqg39hZ5CS5tO9FXfxm5vxrfxYTwdvdhGex/bB/sqcTk1ZveS7jnazLL9qMm5qFHxAx6rKAqKc2xTNoWYLNS0GdhNVdi+gavYTjRWXUmo/+gAK/WRTo314Jx3EdFXfIvYW39N1OVfx3XOp8d7WEKICSBiAjYtez5qUhbZtZs4a146r28txbJtlsVUgGUOKWVAURSITcajtY5xwBZK5RzsCpvtbxv4oA5WYyUE2idNwBbriGFx6gI+rNxOwPxoFVTp2HhttdZR46sO3dje94/92qZ2wCbWV4XmyUVRFFxn3ojtawmlmwzgomU5uJwai2cm43T0DMjMsv3Yfu+w0yFPp6XNwDH9DKzqY3hiQ5veR7N59s6jtXx4oJpLlud06zk3FkJNbo+gZU/sH1lCTARd+9hGuMpmW2Y4hhNWRmnnSv3k+SxRnFGh/WwxSeM9FCHEBBAxAZuiKDgWXIZVd5JPLYLkeBcxbp3kxn0oMUlDqs4H4EhIJWXMAzYvDl0lOWHglMXAgbdpfeTrvZZ7743V8SU80StEnursrDNoN9rZVbO367bO/YdWax2lreUopgNva9+1ceqafMQr7eiGt6uCqJZagD5jFYHdr2C1NfQ7htgoB/9xy3I+densXu8PHt+K4ooeVErfYKnJedi+FhK0dhSgfpRSIlvbg/z5pYPkpMZwzTljv+/RrDgEloGe039VNyHEyGmp00BRRhSwGaV7aX3oc7S/8b+Y9WVhHN3ImGV7UZNyJLgRQkxZEROwAThmnY3iikU9/Ab/58YlfOPaOaekQw5tqGpcKslaK5V1YxuwZXiie+3jdSrbCBDY9jRYBsaJ7YM6t1l1DJzRk6oYw+ykGSS7Pd3SIpWOCop2Sw1lrRW4zCSaWgN9nqO2yUem1gjQreWDa+XHwDa7env1Jzs1lvjo7iWezfpS2l/5L4zDG4idswpFC1+bBjUlDwClsYT4GCf1zaOTEvnYa4dpbQ/y2avmjfnqGnRcFVd1tIxZY/7cQkw1isOF6skNfVcMk1m2H1AwinfgffLfaX/115i1xeEb5DDYhh+z4jDaBE+HFEKIkYiogE3RnTjmXoBRtIN0Rxt5ZhGYQfRhVFBS41JwE6Cxvv8VlnCqqGsb1P614MF3sL2N4IzGKB5kwFZ9DC1t+oTbcN0fVVE5K3MlhxuOUuMNVU5U3HEoUfEEa4opb60gTkmhsZ8VqJqmdvKcjQBono+qh6nxqTjmXUzw0IYhXSm2mmtof+u3eJ/8PkbFQZwrP07y5Z8b3gvsQ+c4zdqTJMW5RqXoyLZD1Xywv4qrzy4gPyMu7OcfDLNsP1rGrD6b3AohwktLm4FZfRzbtob1eLPuJKonh9hP3Ydz6VqMsv14//FDvC//oqutzFgzKw7LSr0QYsqLuF//jvkXg6IS6KigpETFo6UP/Qq90lnav6UW0xrel9dQ+INmqFrhAPvXbCNAYOcLaJmFOOdfjFl5GMvX0v9jAu1YDaWTZv/aqVZlLkdBYVPHKpuiKGhpM6iuP0rACuJxptLYGsDqoz1DXZOPAnczSnQiiju2232uZdeAw43/w78POA7L24Tvvb/Q9vf1GMe3hCoW3vRzXEuvRnWEN+BQnFEo8elYdSfxxLvDnhLZ7A3w8CuHyE+P46qz+q+satadxCjd2+8xw2F5m7DqS+SquBBjSEufAcF2rMaKIT/Wtm2s2mLU5HwUdyyulR8j9lP34lxxPWbVUbxP/xjvS/dhB8YuawVCaZpoOlpm72nrQggxFURcwKbGJKHPWEnw0LsYJ3ehFyxHUYc+zM7m2Um0UNs4ulX4AKrqvdhAZkr/AVvwwNvY3kacy69FL1gOto1ZvLPfx5i1RWDboS/jSSbJncj85EI+qNiC2bHZXU2fSXmgCYCs6ExMy6a1vff2DKGUyIZee/Mo7licS67GPLkLo/xgj/tt08AoP4Bv019p+9udBPe/iaPwPGJuugfXmZ/oEQCGk5aci1l3Ek+ci4YwVom0bZtHXjlEu9/gs1fP7bWvHIAd8OJ77y94//FD2l+8j+CJbWEbA4DZUc5fz5ar4kKMlc6LetYw0iJtbyO2rwWtI2UbQv0RXcuuIfaT9+I84wbM0r34N/01bOMdDLN0H1pGoazUCyGmtIgL2ACcCy8PlXY3/OjTB9cs+3SdxSs8aisVY1B4pLyzQqSn75TIj1bX5qBnzUVNyUeJ8WAU7+j33GbVUQC01KEVXpkozso6g6ZAC/vrDwGhHx3lTh0Vhdz40J69vtIi65u8JJr1qEk5vd7vXHApSowH/+bHQ1eQm6oI7Hsd78u/pPXhr9L+/M8I7n0dPW8xMTf8P9znfWZMNraryXnYzdWkxNi0+82wNc/efKCKbYdqWHfuNHJSewactm0TPLaZtse/R3DfGzjmXoSaNg3fm7/BrDkRljFAR/81Vwxqct7ABwshwkJJSAdXzLAKj1h1ob1qvf3NKs4oXEuuwrn4KoKHNmCc3DXisQ5qTG0NWA2lE76cvxBCjFTf5ffGkZY6DS19FlZT5fCbFbtiwOHGo7VSNQYBW0WtF0WB9H4CtuCBt7Dbm3Be8mUglP6n5y8leGgDtuHv8wqiWXUMNSFjVFd8xtPC5LnEOWJ5v3wLC1PmoaVOo8Klk65GkRwfWrFsbPWTl959L1abL0hssB4NE83Te8Cm6E5cK67D985DtD36zdDeQULN1R2zzkbLWYCeNRfFGTWqr/F0WkcT+AwltMeyvtlHdi8B1lA0tvp59NXDTM+KZ82ZPX90WU2V+DY+glm2DzWlgKg130BLnYblbcL79I9pf/mXRF/3A9TY/huVD8S2bczSfejZ84a1Oi6EGB5FUdHSpg8rYDNrTwKg9XORxbl8HUbxDnzv/pGYG34y6g2fO1fqNdm/JoSY4iL215T74i8RdfV3h92sWFEU1LhU0hzeMSntX1HvJTUxqs9qfLbhD62uZc1FPyUI1QuWgRnALN3f++NsG6v62KQq5386TdVYlbmCvXUHaPI3ozhcVEa5yAyYJHb0KmvspVJkbaOPLD0U8PSWEtlJn3UOesFy1JR8XOd8mpgbf0bsJ3+O+9xbcRQsG/NgDT66ip1shnrNjbTwiG3bPPzyIQKGxWevmot2SqBkGwH8256m7cn/wKw+juvsTxN97Q9CZcABNTqBqDX/B9sI0P7yLwdsOD4Qq7EC29s4qXomCTFRaGkzsOrLhvx3bNWdRIlP7/fzUNEcuC/4PHZ7M773HxvpUAdklO5FiYrvVgFYCCGmoogN2NRYT5+rJoM+R1wKqXrbmJT2r6hr6zcdMrj/Lez2Zpwrrut2u5ZVCM4ogkW9V4u0W2pC+womYcGRU52VtRLLtthcuY3WQBtNKmQ0N5EQEyqn31tKZGdJf1tRURMz+zy3oqpEXfY1otd8E+f8S1AT0kftdQxWqEhKHDHtlcDIe7G9s7OcnUdr+djq6d0at1vtzXif+ymBbU+jFywn5saf4lxwSY+VL82TTdSlX8FqKKP9jQeH1TzXtgyM4h343/8LgKQxCTEOQt8V9pBTnM3aYrR+Lnx1nT+1AOfSqzGOvDdgOv9I2LYV2r+WPX9SVUcWQojhmNSfgkpcCgm0UFnfNqrPY1oWVfXePguO2EE/gV0vomXPR8/oXulKUXX0vMWYJ3di91LNsjO1ZbIHbOnRqcxMnMb75R9S2loOQFZ7O2pLFXHRDhpbewY0dU3tZGkNoavCYeyTNhYURUFNzkNvLgs1zx5mLzbLsnnqnWM8/Moh5hUkccnKj35wWc3VeJ/5CVZ9Ge5Lv0bUxV9CjU7s81x6zgJc59yCWbIb/wd/G9Tz27aNWX08VGHzL9+k/ZX/wqorwbni+q7CP0KIsaOlhfY6DyUt0g54sVtqUFP6ryrbybn0GlRPLr53/4Ttax3WOAdi1Z7E9rVIOX8hhCBC97CFixqXgsMOYHhbaPcbRLmG/3Jt20bpoyF2bZMPw7T77MEW3P8mdnszruXX9nq/nr8M4+gHmFVHuqVLQkfDbN05JVJCzs48g4cPPM7bpe8BkOk3sKqOkRjr7jVlsKbJR6HeiJ4yMVdytJR8zD2vkhSrD2uFzesL8tvn9rP7WB2rF2dy86WFXU3bzdoi2l+6H9syib76TrT0mYM6p3PehVhNlQT3vIIan45zwSU9jrH9bViNFRjlBzCOvB8qIa7p6PlLQ/sCcxeiqJP6o0WIiKW4YlATM4fUQNusKwH637/W7Tk0HfcFn8P7zx/je/9Roi66Y1hj7Y9RtA0UFS1vUdjPLYQQE82k/lWldFzh96itVNZ7mZYZP6zzbNhVzj82HOdfP7ao13NU1IZSLjN76cHWtbqWswAto/d+cnruQlB1jOIdPQO26mNoqdOGvZdvIlmatpC/H36GPbX7SXDGE+Now6w+SmLs8l73sDU3NuFRWydsMKsm54FlMCPGS8MQV9jKa9v49VO7qW3yccvlhVywJKvrgoJRuo/2136N4ooheu16tMSsIZ3bdeaN2M3V+Dc9Cgpg21iNFaH/ayjHbm/qOlbLmI3rvMtxTF856gUIhBCDo6bNwDy5q98LjaeyajsqRA5yhQ1CF5ycy64hsO2fBKetwDFt+bDH2xvjxDa0zEJUd9zABwshxCQ3qVMiO0v7J3cEbMNh2zavbimhqTXAfX/byYmK5h7HVHSU9M/qZYUtuP8NbF9Ln6trECqZrGXPxSjajn1Kg2jbCGDVnpz06ZCdnJqTlRlLAciJy0JLm4FZdZykOGevKZFKUyh1UvMMvO8iEnUWHilwNw5phW3H4Rr+78NbafcbfOeTS7lwaXbXj7Lg0U20v3w/alwK0ev+Y8jBGoT2/Lkv+iKqJw//e3/B//6jBI9swjb8aLkLcJ5xA1GXfZ2YT91H9DX/hnPuBRKsCRFBtPSZ2L4WrKbBNdA264pDxT36SZnujXPpVajJ+fg3/hnL1zKMkfYxnoZyrMZy9DAHgUIIMVFN6hW2rl5sWuuwC48UVbZQVtvGVWfls3l/Fff9bSffumlJt5W2ijovCTFOot3d91HZQR+BXS+FVtcGSEnT85eFvvQaytE82UDHVU/bnNQVIk93dtZKNpRtIjs2Ey0tmUDJHpJTobktgGFaXY2gbdsmylsJbibuCltCBmhOMtV66pszB7wabtk2z71XxDMbT1CQEcdXr1+IJ97ddX9g98v4P/gbWmYhUZf964iCKMXhIvqaf8OsLUKNTwsVSRnElXohxPjTs+fjB8ySvYO6aGPVnRxWz0RF7UyNvAv/xkeI6mhZM1JG0TYA9AIJ2IQQAib5CpvijAZXDNlRPqoahhewvbenAoeucsWZedz5qaVEu3Xu+9tOiio/WmmrqGvrdf9a8Mj7A66uddLzlwAffVHB1Ck4cqq8uBxum3cT5+ecjZYeqnaWpdRgEwraOrX5DFKpx1BdKCPsGzZeFFVFTc4h2azBHxy4efaTbx3jmY0nOHtBButvXtYVrFm+Fnzv/AH/B39Dn7aCqCu+FZYVL8XhQs8sRI1JkmBNiAlEjU9FSUjHKN074LG2GcSqL+/qDTlUWnIuzmXrMI5/iFFxaFjnOJ1RtB01bTpqTFJYzieEEBPdpA7YILTKlu7wDmuFLWhYbN5fxbLZqUS7HaQkRHUFbff+NRS02bZNeZ231/1rZskelLjUQRV8UGOSUFOndyuTbFYfQ4lNHnKaykR3RsYyEl0JXX3CkgOh1MdT97HVNrWTqTUQjM2c0MGElpxHnK8SsKlv7jst0rJsNu6pYHlhKp+9ai5Oh4ZtmQT2vk7b4+sJHt6Ic8lVuC/+MoruHLsXIISISHrOAszyg9hGz/2/p7IaykOZHMNYYevkXHQ5OKIIHnx32OfoGk9rHVbNCfSCFSM+lxBCTBaTP2CLTSFRaaGywYt1yv6wwdh1tJY2n8E5CzO6bjs9aNt1rI52v9Fjhc22TIzyg+hDaB6sFyzDqjmB1RZqBm1WHZtSq2unC1U7yyLOWwrQbR9bbUM7WVrjhE2H7KQm56EZ7SSpbf3uYzte0Uxre5CVc9JQFAWjbD/ep36I//2/oKUUEP3x/8R1xg09+qsJIaYmPWchmAHMyiP9HtdZcERLHt4KG4Ciu3DMPBPj+BbswMj6nhonQlkmjmnLRnQeIYSYTCb9rzslPpUYo4lA0Oy1+XJ/Nu6pICnOxdzsmK6yx9ARtH0yFLQ98NQegB492KyaExBsR8uZN+jn0wtCBTeM4h1YbQ3YbfUdaYFTl5o2A0djEWB3K+3fXFtNtBrAnT78HxmRoLOMdrZWT31L35Uidx2tRVUU5qfatL/2AO0v3INt+HBf9jWirvw2WlL2WA1ZCDEBaFlzQNUGTIs064pBd6EkpI3o+RyFq8EMEDy6eUTnMYq2oXpyQnt8hRBCAFMgYFNjU1BtgzjFN6RKkY2tfvYcr+PsBRkYO5/H+4+7sLyNXfenJIaCNk+8C4Cs01IijdJ9gIKeNfiATU3MCu07KNo+Jfev9UZLn4HibyVVa+22wmbUnQSY8AGb6skFFHL1+n5TIncfq+Pq9DKsZ76PUbIb54rribnh/+EoWD6hU0KFEKNDcbjRMmZjDhCwWXUlqMm5KMrIfg6oqdNQPTkEDw0/LdLyNmFWHJZiI0IIcZrJH7DFd5T214ZW2n/Tvkpsm1DAVrwDbBOjeGe3Y1ISo1h/8zI+f/U8kuJc3e4zy/ahphaguGMH/ZyKoqDnL8UsP4BZshdUfUh9cSajzhXGeTEN3QI2vaWjpP8EX1lSHC7UxAwKXE009LHCVt/so6K6idXm+6gp+cR84qe4ll0je9WEEP3SchZg1Zd0u9h4Ktu2MOtOjigdspOiKDgKV2PVnOiWkTIUoT3cNvo02b8mhBCnmvQBmxIbap6d7hx84RHbtnlvTyUzsuNJd3ixGsqA7hUcO3ni3Zy1oHvqhh1ox6w6hp49+NW1TnrBcrBMgoc3oKbkoWiOgR80iamJ2eBwM91Z163oSEx7Na1q3KTo/6Um55Gl9r3CtvtYHUudRThML64V16NO0KqYQoixpecsAOhzlc1uroGgDzVl+AVHTuWYdTao+rBX2YyibShxqRN+b7IQQoTbpA/YOnux5UX7qBxkaf+iyhbKa9s4d2EmxsldAKGVr7L92IH2AR9vVhwC20QbQsGRTlraDBR3HFjmlE+HhFDpey11GjlKddceRNu2STJraXOnj/PowkNNziOeFtpbejZlh9D+tQtjDqMkZqFlzR3j0QkhJio1ORclKh6jpPeAzawbecGRUynuWPSCZaGWNmZwSI+1A17Msv3o0yTNWwghTjfpAzbF4UJxx5Huah/0CtvGjt5rK+ekY5zciZqQgWPxlWCZXQFcf4yyfaA5B1XOv8d4VRU9P1R8RAK2EC1tBh6zhrbWNgBaWttJUxsx4jLHeWTh0Vl4JMpbjn1aJdNA0KSl5AjZSg3OeRfJDxkhxKApioqWswCzbB+2bfW436o9CYqGmjRwc+3BcsxZDf42jKLtQ3qccXIXWCYOSYcUQogeJn3ABqDEpeJRW6hr8hE0zH6PDRomH+6vYvnsVKLUIGb5QbT8JaGVr6j4QX0JmWX70DJnD3uPkV54HkqMJ1TlS6Clz0DFwmNUhap9lp9EU2y05NzxHlpYdPY/SrfraPN1b5598GQDqxwHsDQXjtnnjMfwhBATmJ6zANvX0lW+/1Rm3UnUpMyw7ofVsuehxCYPuSebcWIbSnQiatr0sI1FCCEmiykRsKlxKcSazdhAVUP/KY07j9Z19F7LxCjbD5aBnre4a+XLKNndb6qH1daA1VA+pP5rp9MzZhF78/1TrmF2X9SOlcZ8vZbGtgDtlUUARGcUjN+gwkiNTiDojCNbr+/WugDgwOESljmLcMw6C8UZNU4jFEJMVFrHPjajZE+P+6zaYtQwpUN2UhQVx+xzMcv2Y7XUDuoxtuHHKNmNXrBsxNUqhRBiMpoSn4xqXApOfyMKFlUDVIrcuLuj91p+EubJneCMQsuYBXQUBAn6MMv29/l4s2wfAFrO8AM20Z0aFU8wKpkCvYbGFj9mfSmGrZKYFZ6N8pHASsghW2ugvvmjSpG2beMs2oRDMXEtuHQcRyeEmKjUqHjUlPwehUcsbyN2exNamAqOnMpReB4AwUMbBnW8UboXjIBUhxRCiD5MiYBNiUtBsU3ilfZ+S/s3tPjZeyLUe01RbIyTu9BzFqKoOgBa9lxwuPtNizRK96G446TKVZjZKdMp0GtpbPXjaK2gxk4gJto93sMKG0dqPhlaIw1NbV23ldW0sIT9tMRNQ/NM7PYFQojxo+cswKw61q1oltXRy7IzJTuc1LgUtJz5BA9vxLZ67p07nXFiG7hi0DJnh30sQggxGUyJgE2NC5X2z4/191t45IOO3mvnLszEqinCbm9Gz1/Sdb+iOdBzF2EU7+j1S8i2bcyy/WjZ8yWtI8xcmTNJVL1466qJ9VXRqKWO95DCKipjOppiEzylf1HJjg9I0VqJXiSra0KI4dNyFoR6iZZ/lB1i1oYCNm0UAjYAR+Fq7NY6zPK+M1IAbMvAKN6Jnr+k6+KoEEKI7qZEVNFZ2r8g1t/nCptt22zcU8HMnATSPdEYJ3eCoqDnLup2nF6wDLu9GbP6WI9zWA2l2O1N6JIOGXburFBaqqvmALF2K96oyVHSv5OeGvrRpDaWdt0WW/IerUSTMPeM8RqWEGIS0NJngcONeUp5f6vuZCj7ZJR6WeoFS1FcsQMWHzHLD0LAi6NA0iGFEKIvU+JyltLRaDjT7eO5E83c+eD7JMQ6SYhxkRDjJCHGCQpU1Hn5zBWhyoxG8S609Fko7thu59LzFoOqYRRtQ+/Y29bJLA1dSdSG0TBb9E9LziNoa2Q2hJqXm/HhK0MdCZT4NAI4cLdVANBSXUaeWUxRymoy5aqzEGIEFE1Hy5yDUboX27ZRFAWzrjhs/dd6f04H+qyzCO5/C9Pb0udxxoltoLtk37cQQvRjSqywKboTJTqR2UkGl52Ry8ycBJy6RlW9ly0Hq3l64wme3nCCKJfGisK0UKXHumK0vMU9z+WMQsueh3FiW4+eWUbZPtSEDNSOAFGEj6Lp1GhpJBtVADhSJkdJ/06KotLoSCUpEHp9tR++jI1CwlJJhxRCjJyeuwC7pQa7uQo70I7dVI06CgVHTuWYsxosg9a97/R6v20aGEXb0fMWhbW1gBBCTDZT5tK9EpeCy9/AjZfN6nGfYVo0twXQNJVot07gQKg5tp63pNdz6QXL8W/4E1ZDKZonFDjYZhCz4mBXdSwRfo3ubLK8FbRZTuJTJ1dKJIA3OpOMwG6sgI+4ii0csAo4c5oUrxFCjJyesxA/YJTsRU3JB+xR27/WSfPkoqZOo2XXG+jx07AayrEayz/6/41VYJuhCsxCCCH6NGUCNjUuBbPqaK/36ZqKJ/6jioNG8U6UuBTUpN7T7vT8Jfg3KBgntncFbGbVMTACaCPovyb61x6XB96tVJhJpCZOvp5kZnwO7qZttG55Bpftoyb9TFRFGe9hCSEmATUhHSUuFaN0T9cXf7h7sPXGUbga/8Y/E3jqB6EbFBUlPg0tMRM9fxmqJwd9+spRH4cQQkxkUyhgS8U49iG2ZaKoWp/H2UYAs2w/jjmrUfr4saxGJ6Kmz8Ao2o5r+Tqgo/+aoqJnzRmV8QuwkqdDFZSbicyJnzwl/TtpKXlQAvb+V6k0Esmau3S8hySEmET03IUED7+H4opBccWixCSN+nM6Cs8lNlqnzXKhJmaFAkfNMerPK4QQk8mU2MMGoZRIbAu7rb7f48zy/WAGupXz742jYBlWXTFWSw0Q6r+mpk1HcUaHa8jiNNFJKTztXc4O5hPlmnzXGmIyCjBtBdU2eS9QyPzpshdSCBE+Ws4CMPwYx7egpuT3eVEynBTNQcLKK3FMPwPNkyPBmhBCDMOUCdg6e7FZLbX9HmcU7wpVrMos7Pe4zpx7o2gHtr8Nq/YEuqRDjqqkOBdv+eZjxWeO91BGRVJiLFVmAj7bQVPaskkZlAohxo+eNRcUDczgqDTMFkIIMTqmzC/Czl5sdnMNZM3t9RjbtjFO7kLPWTDgVUA1IR01KQejaBtKrAdsW8oSj7LEWBcAKQmTLx0SICHGyZ99S9EwmLd0cgalQojxozij0DJmYlYcCqVgCyGEmBCmzAqbEuMBRcFq7XuFzaovwW6rD/VaGwS9YClm5WGMox+Aw42WNj1cwxW9+Chgm3wFRwBUVaHMPYsdgWksmpky3sMRQkxCWs5CgI5KkUIIISaCKbPCpmg6SnQSVnNNn8cYxTsBeu2/1ht92nICO57DOLEVLW8xijQ4HlUup8Znr5pLYW7ieA9l1CQnuNE1hQyP7IUUQoSfc8GlqEmZaIm9V0EWQggReaZUhKHGp2K31vV5v3FyJ2rqdNTohMGdLzkfJTYZu7UOPWdBuIYp+nHOwsmdKnjLZbMxLXvgA4UQYhgUhwuH9D0TQogJZcqkREKoUqRZX4J/x3MEj36AWX0c29eKbdtY3ias6hPo+YNbXQNQFAW9YBkAWva80Rq2mEKyU2PJS48b72EIIYQQQogIMaVW2PS8JZglewlsear7Hc4oFFcsYKPnLRnSOZ1LrkLz5KJKeokQQgghhBAizKZUwOaYvhLH9JXYQT9WSw1WczV2czVWx/8padOHXOpYjU5EnbN6lEYshBBCCCGEmMqmVMDWSXG40Dw5aJ6c8R6KEEIIIYQQQvRpSu1hE0IIIYQQQoiJRAI2IYQQQgghhIhQErAJIYQQQgghRISSgE0IIYQQQgghIpQEbEIIIYQQQggRoSRgE0IIIYQQQogIJQGbEEIIIYQQQkQoCdiEEEIIIYQQIkJJwCaEEEIIIYQQEUoCNiGEEEIIIYSIUBKwCSGEEEIIIUSEkoBNCCGEEEIIISKUBGxCCCGEEEIIEaEkYBNCCCGEEEKICCUBmxBCCCGEEEJEKAnYhBBCCCGEECJCScAmhBBCCCGEEBFKAjYhhBBCCCGEiFASsAkhhBBCCCFEhNLHewAAqqqM9xAmFZnP8JM5DS+Zz/CTOQ0vmc/wkvkMP5nT8JL5DD+Z08EbaK4U27btMRqLEEIIIYQQQoghkJRIIYQQQgghhIhQErAJIYQQQgghRISSgE0IIYQQQgghIpQEbEIIIYQQQggRoSRgE0IIIYQQQogIJQGbEEIIIYQQQkQoCdiEEEIIIYQQIkJJwCaEEEIIIYQQEUoCNiGEEEIIIYSIUKMSsD3wwANcddVVXHXVVdxzzz0AvP/++6xdu5bLLruMX/ziFz0ec+edd/KPf/yj69/l5eXcfPPNrFmzhi996Uu0tbWNxlAnjHDMKcATTzzB+vXrx2TMkSwc87lt2zY+/vGPs27dOm677TbKysrGbPyRJhzzuXXrVq6//nrWrl3LF7/4RZqamsZs/JEoXH/zAPv372fBggWjPuZIFo75/Oc//8m5557LunXrWLduXa+PmUrCMafV1dV84Qtf4Nprr+Wmm26itLR0zMYfaUY6n3V1dV3vzXXr1nHRRRexdOnSMX0NkSQc78/S0lJuvvlm1q1bxy233DKlv+chPHO6e/duPvaxj7F27VruuOMOampqxmz8E5odZu+9955944032n6/3w4EAvatt95qP/fcc/b5559vnzx50g4Gg/btt99uv/3227Zt23ZlZaV9xx132IsWLbKfeuqprvN84QtfsJ9//nnbtm37gQcesO+5555wD3XCCMec+nw+++c//7m9ZMkS+7vf/e54vpxxF6736IUXXmgfOHDAtm3bfuKJJ+wvfvGL4/J6xlu45vOSSy6xjxw5Ytu2bf/85z+377vvvnF5PZEgXHNq27bt9Xrtm266yZ49e/Z4vJSIEK75/PGPf2w/99xz4/UyIkq45vS2226zH3vsMdu2bfuxxx6zv/71r4/Hyxl34fybt23bNk3T/vSnP20/++yzY/1SIkK45vPb3/62/eijj9q2bdsPP/yw/a1vfWtcXk8kCMecWpZln3/++famTZts27btF154wb7jjjvG7TVNJGFfYUtNTWX9+vU4nU4cDgczZsygqKiI/Px8cnNz0XWdtWvX8vLLLwPw3HPPcfHFF3PFFVd0nSMYDLJlyxYuv/xyAK6//vqu46eicMzpli1bsCyL73znO+P1MiJGOOYzEAjw9a9/nTlz5gBQWFhIRUXFuLye8RaO+QR48cUXmTlzJsFgkKqqKuLj48fj5USEcM0pwN13381tt9021i8hooRrPvfs2cM///lP1q5dy7e//e0pvQocjjmtr6/n4MGD3HTTTQB87GMf4xvf+MZ4vJxxF86/eYCnnnqKqKgo1q5dO5YvI2KEaz4ty6K1tRWA9vZ23G73mL+WSBGOOW1oaMDn87Fq1SoALrzwQjZu3EggEBiX1zSRhD1gmzVrFkuWLAGgqKiIl156CUVRSE1N7TomLS2NqqoqAD73uc9xww03dDtHQ0MDsbGx6LoOhN4kncdPReGY03PPPZc777xzSn/YdArHfDqdTtatWweEPtAfeOABLrnkkrF5AREmHPMJ4HA4OHToEOeffz6bN2/mqquuGpPxR6Jwzekbb7yBz+djzZo1YzLuSBWu+UxNTeXLX/4yzz77LJmZmfz4xz8ek/FHonDMaUlJCVlZWdx999187GMf41//9V9xOBxj9hoiSbjeowCmafK///u/fOtb3xr1cUeqcM3n17/+df70pz9x3nnn8Yc//IHPf/7zYzL+SBSOOU1KSiI6OpqNGzcC8MILLxAMBmloaBibFzGBjVrRkSNHjnD77bdz5513kpubi6IoXffZtt3t36fr7f7+jp8qRjKnoqdwzGcgEODb3/42hmFwxx13jOZwI1445rOwsJD333+fL3/5y3zzm98czeFOCCOZ05qaGh588EG+//3vj8VQJ4SRvkf/+7//m+XLl6MoCp/73OfYsGHDaA854o1kTg3DYP/+/axatYqnnnqKiy++eMrvsQ7H5+iGDRsoKCigsLBwNIc6IYx0Pr/73e/y4x//mA0bNvCjH/2Ir371q9i2PdrDjmgjmVNFUfjVr37Fb37zG6699lpaWlpITEycshdqhmJUArZt27bxmc98hm9961tcd911ZGRkdNtUWFNTQ1paWp+P93g8tLS0YJrmoI6fCkY6p6K7cMxnW1sbn/vc5zAMgwcffHBKf+CMdD79fj+vv/5617+vueYaDh06NKpjjnQjndO3336bxsbGrg3zAOvWretK75lqRjqfLS0t/OlPf+r6t23baJo2mkOOeCOd09TUVGJiYrjwwgsBuPrqq9m9e/eojztShet7/vXXX+fKK68czaFOCCOdz/r6eo4fP96VPXP55ZdTU1MzpVeDwvEe1XWdRx55hKeffpprrrkGy7JITEwc5ZFPfGEP2CoqKvjKV77Cvffe25XStHjxYk6cOEFxcTGmafL888+zevXqPs/hcDhYsWIFL774IgBPP/10v8dPduGYU/GRcM3nd77zHfLz8/nlL3+J0+kci6FHpHDMp67r/OhHP2Lv3r0AvPTSSyxbtmxMxh+JwjGnN9xwA6+//jrPPPMMzzzzDADPPPMMsbGxY/IaIkk45jM6Oprf//737Nq1C4C//OUvXHrppWMy/kgUjjnNy8sjIyODd955B4C33nqL+fPnj8n4I004v+d37tzJihUrRnvIES0c85mUlITL5WLr1q1AKFiJiYnB4/GMyWuINOF6j/7bv/1b14WZP/7xj6xZswZVlS5jA9HDfcKHHnoIv9/P3Xff3XXbTTfdxN13383XvvY1/H4/559//oB7Kn74wx+yfv16HnzwQTIzM7n//vvDPdQJI1xzKkLCMZ/79+/njTfeYObMmVx33XVAKHf7d7/73aiPP9KEYz41TeMXv/gFP/jBDzBNk/T0dH7yk5+MxfAjkvzNh1e43qO//OUvueuuu/D5fBQUFHSVtZ6KwvUe/fWvf80Pf/hDfv7znxMbG9vtfFNJOP/mS0pKyMjIGM3hRrxwzKeiKDzwwAP853/+Jz6fj5iYGH7961+PxfAjUrjeo3fddRc//OEPaW9vp7CwcEp/1w+FYk/1ZFwhhBBCCCGEiFCyBimEEEIIIYQQEUoCNiGEEEIIIYSIUBKwCSGEEEIIIUSEkoBNCCGEEEIIISKUBGxCCCGEEEIIEaEkYBNCCDHuSktLKSws5NOf/nSP+9avX09hYSH19fXDOvcDDzzQ1Zh9/fr1PPTQQz2Oefnll1m5ciWVlZXdbr/nnnv49Kc/jWmaw3ru3hw8eJBzzz03bOcTQggxuUnAJoQQIiK4XC5OnDhBWVlZ121er5ft27eP6LybN2/GMIx+j1mzZg2XXnop3/ve9+jsdrN161aeeeYZ7rvvPjRNG9EYAAzD4E9/+hOf/exnaWtrG/H5hBBCTA1hb5wthBBCDIemaVxxxRU899xzfPGLXwTg1Vdf5eKLL+YPf/gDAI8//jiPPPIIqqqSkpLC97//faZNm8b69euJjY3l0KFDVFZWUlhYyM9+9jOefvpp9u7dyz333NMVdO3YsYObbrqJ2tpaZs2axX333Ud0dDT/8R//wXXXXcdjjz3Gtddey/r16/npT39Keno627dv595776W9vR1VVfnqV7/KhRdeiNfr5a677qK4uJjGxkZiYmK49957mT59OrfccgsJCQkcP36cT37ykyxevJhDhw7xwAMPcPvtt4/bPAshhJhYZIVNCCFExLj22mt55plnuv799NNPc9111wHwwQcf8Pvf/56HH36YZ599lquvvpqvfOUrXStie/fu5aGHHuLFF1+krKyMl19+mZtvvpkFCxZw5513cumllwJQVVXFH//4R1555RWqqqp49dVXAYiOjub+++/nV7/6Fd/73ve44oorWL16NU1NTXzve9/jnnvu4Z///Cf/8z//w1133UV5eTnvvvsu8fHxPP7447zyyissWLCARx99tGv88fHxvPjii9xyyy0sWrSIn/70p6Smpo7VdAohhJgEZIVNCCFExFiwYAGaprF3716Sk5Npa2tj9uzZAGzYsIErr7wSj8cDwPXXX89PfvITSktLATjvvPNwOp0AzJ49m6ampl6f45JLLiEqKgqAWbNmddsbN3/+fD71qU/x+uuvc//99wOwc+dOampq+MpXvtJ1nKIoHDp0iDVr1pCbm8sjjzxCcXExH374IUuXLu06bsWKFeGaGiGEEFOUBGxCCCEiyjXXXMOzzz6Lx+Nh3bp1XbcritLjWNu2u/anud3ubsd2rrydTtf1fo/Lzc0lKyur6zjTNJkxYwZPPPFE1zFVVVV4PB4ee+wx/v73v3PzzTezdu1aEhMTuwJICK3aCSGEECMhKZFCCCEiyrp163j55Zd58cUXufrqq7tuP++883jxxRe7VsSeeuopEhMTyc/P7/d8mqYNWHSkP0uWLKG4uJgtW7YAcODAAS6//HKqqqrYuHEj1113HTfccAPTpk3jzTffDGtFSSGEEEJW2IQQQkSU9PR0ZsyYQVxcHImJiV23n3nmmXzmM5/htttuw7IsPB4Pv/nNb1DV/q89XnTRRdx///0Eg8Fhjcfj8fCrX/2Ke+65B7/fj23b3HPPPeTk5HD77bfzgx/8gCeffBIIBXeHDx8e1vMIIYQQvVHsvnJGhBBCCCGEEEKMK0mJFEIIIYQQQogIJQGbEEIIIYQQQkQoCdiEEEIIIYQQIkJJwCaEEEIIIYQQEUoCNiGEEEIIIYSIUBKwCSGEEEIIIUSEkoBNCCGEEEIIISKUBGxCCCGEEEIIEaH+PynWXyO3vIKjAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x648 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Merge DAX cumulative log returns with the cumulative log returns of the value weighted portfolios and plot\n",
    "dax = dax.rename(columns={'cumulative_log_return': 'cumulative_log_return_dax'})\n",
    "long_short_dax_vw = pd.merge(long_short_vw,dax[[\"cumulative_log_return_dax\"]],left_index=True, right_index=True)   \n",
    "long_short_dax_vw[[\"cumulative_log_return_long_vw\", \"cumulative_log_return_short_vw\",\"cumulative_log_return_dax\"]].plot(grid=True,figsize=(15, 9));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save results to compare with the other ML models\n",
    "long_short_dax_ew.to_csv(r'C:\\Users\\krist\\Documents\\Data\\cumulative_log_returns\\rf_ew.csv')\n",
    "long_short_dax_vw.to_csv(r'C:\\Users\\krist\\Documents\\Data\\cumulative_log_returns\\rf_vw.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}