RMaFI

Solutions for exercises from John C. Hull's Risk Management and Financial Institutions using Python and Pandas.

So far I tackled Chapters 10 Volatility, 11 Correlations and Copulas, 13 Market Risk VaR: The Historical Simulation Approach, 14 Market Risk VaR: The Model Building Approach, 18 Fundamental Review of the Trading Book, and 19 Estimating Default Probabilities.

Requirements

For the majority of problems the following will suffice:

python3 -m pip install -r requirements.txt

For one problem in Chapter 11 Copulas and Correlations and a few problems relying on the Extreme Value Theory in Chapter 13 The Historical Simulation Approach, where we need to solve optimization problems, TensorFlow 2.x will be needed:

python3 -m pip install -r requirements_extra.txt

How to install using virtualenv

If you want to avoid mixing up packages in your global python installation, you can prepare the environment as follows:

python3 -m venv venv
source venv/bin/activate
python3 -m pip install -r requirements.txt
python3 -m pip install -r requirements_extra.txt
python3 -m pip install jupyter

Data for Chapters 13, 14, and 18

The spreadsheet I used for exercises in these chapters can be downloaded from this page. I used VaRExampleRMFI3eHistoricalSimulation.xls for all these Chapters. Please remove the first row before feeding it to the code (for the spreadsheet in this repository I already did it).

Code in ch14/ replicates the functionality in VaRExampleRMFI3eModelBuilding.xls.

How to run

• Chapter 10: Volatility Using Jupiter.

  jupyter notebook ch10.ipynb
  

  A full run of the notebook can be seen here.

• Chapter 11: Correlations and Copulas

  jupyter notebook ch11.ipynb
  

  A full run of the notebook can be seen here.

  You can also run the notebook in Google cloud. This way you don't need to install anything locally. This takes just a few seconds:

  1. Go to Google Colaboratory in your browser
  2. In the modal window that appears select GitHub
  3. Enter the URL of this repository's notebook: https://github.com/ilchen/RMaFI/blob/master/ch11.ipynb
  4. Click the search icon
  5. Enjoy

  I make heavy use of the SciPy and TensorFlow 2 libraries in these exercises. as they draw heavily on probability theory and solving optimization problems. These exercises are fairly simple from software design point of view, so I limited the implementation to a Jupiter notebook.

  Despite there not being that many challenging exercises concerning the tracking and forecasting variances-covariances, I decided to create required classes to track and visualize correlations in a separate Jupiter notebook:

  jupyter notebook ch11_variance_covariance.ipynb
  

  A full run of the notebook can be seen here.

• Chapter 13: Market Risk VaR: The Historical Simulation Approach

  python3 ch13/hist_simulation.py ./ext/VaRExampleRMFI3eHistoricalSimulation.xls
  

  Or, much better, using Jupiter. For historical simulation exercises that don't utilize the Extreme Value Theory:

  jupyter notebook ch13.ipynb
  

  For exercises that make use of the Extreme Value Theory:

  jupyter notebook ch13_evt.ipynb
  

• Chapter 14: Market Risk VaR: The Model Building Approach

  python3 ch14/model_build.py ./ext/VaRExampleRMFI3eHistoricalSimulation.xls
  

  Or, much better, using Jupiter.

  jupyter notebook ch14.ipynb
  

• Chapter 18: Fundamental Review of the Trading Book:

  python3 ch18/frtb.py ./ext/VaRExampleRMFI3eHistoricalSimulation.xls
  

• Chapter 19: Estimating Default Probabilities

  jupyter notebook ch19.ipynb
  

  A full run of the notebook can be seen here.

  You can also run the notebook in Google cloud. This way you don't need to install anything locally. This takes just a few seconds:

  1. Go to Google Colaboratory in your browser
  2. In the modal window that appears select GitHub
  3. Enter the URL of this repository's notebook: https://github.com/ilchen/RMaFI/blob/master/ch19.ipynb
  4. Click the search icon
  5. Enjoy