Start.R

# Read the dataset
df <- read.csv("C:/Users/sudhi/OneDrive/Desktop/BIS581_FinalProject/Sales.csv", 
               header=TRUE, stringsAsFactors=FALSE);
View(df)

install.packages("tidyverse")
library(tidyverse)

#Checking the structure of the data so that I can figure out the next steps.
str(df)

#Date column is in character format
#So, I'm converting Date to Date type
#Do not execute below statements, as it turned Date rows to NA
  #df$Date <- as.Date(df$Date, format="%m/%d/%Y")
  #str(df)

install.packages("lubridate")
library(lubridate)

# Checking if there are any missing values
missing_values <- sapply(df, is.na)
#missing_values
# Get the number of missing values in each column
sum(missing_values)


# Trim any leading or trailing spaces in the columns
df$Date <- trimws(df$Date)
df$Day	<- trimws(df$Day)
df$Month	<- trimws(df$Month)
df$Year	<- trimws(df$Year)
df$Customer_Age	<- trimws(df$Customer_Age)
df$Age_Group	<- trimws(df$Age_Group)
df$Customer_Gender	<- trimws(df$Customer_Gender)
df$Country	<- trimws(df$Country)
df$State	<- trimws(df$State)
df$Product_Category	<- trimws(df$Product_Category)
df$Sub_Category	<- trimws(df$Sub_Category)
df$Product	<- trimws(df$Product)
df$Order_Quantity	<- trimws(df$Order_Quantity)
df$Unit_Cost	<- trimws(df$Unit_Cost)
df$Unit_Price	<- trimws(df$Unit_Price)
df$Profit	<- trimws(df$Profit)
df$Cost	<- trimws(df$Cost)
df$Revenue <- trimws(df$Revenue)

TrSpaceAfter <- df
View(TrSpaceAfter)

# Convert categorical variables to factors
TrSpaceAfter$Country <- factor(TrSpaceAfter$Country)
TrSpaceAfter$Year <- factor(TrSpaceAfter$Year)
TrSpaceAfter$Customer_Age  <- factor(TrSpaceAfter$Customer_Age)
TrSpaceAfter$Age_Group <- factor(TrSpaceAfter$Age_Group)
TrSpaceAfter$Customer_Gender <- factor(TrSpaceAfter$Customer_Gender)
#TrSpaceAfter$country <- factor(TrSpaceAfter$country)
TrSpaceAfter$State <- factor(TrSpaceAfter$State)
TrSpaceAfter$Sub_Category <- factor(TrSpaceAfter$Sub_Category)
TrSpaceAfter$Product <- factor(TrSpaceAfter$Product)
TrSpaceAfter$Unit_Cost <- factor(TrSpaceAfter$Unit_Cost)
TrSpaceAfter$Unit_Price <- factor(TrSpaceAfter$Unit_Price)

TrSpaceAfter$Month <- factor(TrSpaceAfter$Month)
TrSpaceAfter$Product_Category <- factor(TrSpaceAfter$Product_Category)

factorDf <- TrSpaceAfter
View(factorDf)

factorDf$Date <- as.Date(factorDf$Date, format = "%Y-%m-%d")
str(factorDf)

# Converting required columns into numeric format
factorDf$Day <- as.numeric(factorDf$Day)

factorDf$Customer_Age<- as.numeric(factorDf$Customer_Age)	
factorDf$Order_Quantity<- as.numeric(factorDf$Order_Quantity)	
factorDf$Unit_Cost<- as.numeric(factorDf$Unit_Cost)	
factorDf$Unit_Price<- as.numeric(factorDf$Unit_Price)	
factorDf$Profit<- as.numeric(factorDf$Profit)	
factorDf$Cost<- as.numeric(factorDf$Cost)	
factorDf$Revenue<- as.numeric(factorDf$Revenue)

# Check the structure of the Column dataType after conversion
str(factorDf)
summary(factorDf)
View(factorDf)

#Seperating Age Group and range into two different columns
# Assuming your data frame is named 'df'
factorDf$Age_Range <- gsub(".*\\((.*)\\).*", "\\1", factorDf$Age_Group)

# Remove the content within brackets from 'Age_Group'
factorDf$Age_Group <- gsub("\\(.*\\)", "", factorDf$Age_Group)

# Remove leading and trailing whitespaces in 'Age_Range'
factorDf$Age_Range <- trimws(factorDf$Age_Range)

# Display the modified data frame
View(factorDf)

# Checking if there are any missing values
missing_valuesOfFactorDf <- sapply(factorDf, is.na)
#missing_values
# Get the number of missing values in each column
sum(missing_valuesOfFactorDf)

View(factorDf)

#Start my Visualization from here

#Let's check summary so far.
summary(factorDf)

analysisDf <- factorDf
View(analysisDf)



#Analysis


library(ggplot2)

library(dplyr)

# Install and load the 'viridis' package
install.packages("viridis")
library(viridis)

# Viz 1, creating a Bar plot for Product Category
ggplot(analysisDf, aes(x = Product_Category, fill = Age_Group)) +
  geom_bar(position = "dodge") +
  labs(title = "Product Category Distribution by Age Group",
       x = "Product Category",
       y = "Count") +
  scale_fill_manual(values = c("pink", "beige", "brown", "chocolate", "orange")) +  # Specify your own colors
  theme_minimal()

age_range_colors <- c("<25" = "#FF5F1F", "25-34" = "#FFBF00", "35-64" = "#E3735E", "64+" = "#FA5F55")



# Viz 2, creating a Scatter plot to show difference between Order Quantity vs. Profit
ggplot(analysisDf, aes(x = Order_Quantity, y = Profit, color = Age_Range)) +
  geom_point() +
  labs(title = "Scatter Plot of Order Quantity vs. Profit",
       x = "Order Quantity",
       y = "Profit") +
  scale_color_manual(values = age_range_colors) +
  facet_wrap(~Age_Range, scales = "free")

my_colors <- c("#880808", "#AA4A44", "#E97451", "#F88379", "#E0115F")



# Viz 3, Creating a pie chart for Age Group distribution with count labels
ggplot(analysisDf, aes(x = "", fill = Age_Group)) +
  geom_bar(width = 1, color = "white") +
  geom_text(stat = "count", aes(label = stat(count)), position = position_stack(vjust = 0.5), color = "black", size = 4) +
  scale_fill_manual(values = my_colors) +
  coord_polar("y") +
  labs(title = "Pie Chart of Age Group Distribution with Counts") +
  theme_minimal()

# Viz 4, Faceted bar plot for Country Distribution
ggplot(analysisDf, aes(x = Age_Group, fill = Age_Group)) +
  geom_bar() +
  facet_wrap(~ Country, scales = "free") +
  labs(title = "Country Distribution by Age Group",
       x = "Age Group",
       y = "Count") +
  theme_minimal()

#Viz 5, creating a graph for Average Profit by Age Group
ggplot(df, aes(x = Age_Group, y = Profit, fill = Age_Group)) +
  geom_bar(stat = "summary", fun = "mean") +
  scale_fill_manual(values = c("#0818A8", "#191970", "#5F9EA0", "#6495ED")) +
  labs(title = "Average Profit by Age Group",
       x = "Age Group",
       y = "Average Profit")

View(analysisDf)


newAnalysisDf <- analysisDf


#Analysis



# Descriptive statistics


# Check the lengths of vectors
length(analysisDf$Unit_Price)
length(analysisDf$Gender)


# Descriptive statistics
summary(analysisDf)

# Check for missing values
sum(is.na(analysisDf$Order_Quantity))
sum(is.na(analysisDf$Unit_Price))
sum(is.na(analysisDf$Revenue))
sum(is.na(analysisDf$Age_Group))
sum(is.na(analysisDf$Gender))
sum(is.na(analysisDf$Country))

# Using tapply here to know the sum of order quantity
tapply(analysisDf$Order_Quantity, analysisDf$Age_Group, sum, na.rm = TRUE)

#Countries Revenue sum
tapply(analysisDf$Revenue, analysisDf$Country, sum, na.rm = TRUE)

# Visualize the statistics 
#ggplot(analysisDf, aes(x = Age_Group, y = Order_Quantity)) + 
#  geom_bar(stat = "summary", fun = "sum", fill = "skyblue") +
#  labs(title = "Purchase Quantity by Age Group")

#ggplot(analysisDf, aes(x = Country, y = Revenue)) + 
#  geom_bar(stat = "summary", fun = "sum", fill = "orange") +
#  labs(title = "Total Revenue by Country")


# Linear regression
lm_model <- lm(Revenue ~ Customer_Age + Unit_Price + Order_Quantity, data = analysisDf)
summary(lm_model)


# Chi Square analysis to test the relation between Age_Range and Customer_Gender
chisq.test(table(analysisDf$Age_Range, analysisDf$Customer_Gender))

# Correlation matrix
cor(analysisDf[, c("Order_Quantity", "Unit_Price", "Profit", "Customer_Age")])

install.packages("tinytex")
tinytex::install_tinytex()