# Get the number of rows and columns
total_rows <- nrow(data)
total_columns <- ncol(data)

# Display the results
cat("Total Rows: ", total_rows, "\n")

cat("Total Columns: ", total_columns, "\n")

str(data)

summary(data)

sum(is.na(data))

numeric_cols <- data[, sapply(data, is.numeric)]
cor_matrix <- cor(numeric_cols, use = "complete.obs")
print(cor_matrix)

library(corrplot)
corrplot(cor_matrix, method = "circle", type = "lower")

# Create a bar plot showing the number of accidents by weather condition
ggplot(data, aes(x = as.factor(Wthr_Cond_ID))) + 
  geom_bar(fill = "skyblue") +
  labs(title = "Number of Accidents by Weather Condition", 
       x = "Weather Condition", 
       y = "Number of Accidents") +
  theme_minimal()

# Create a bar plot showing the number of accidents by light condition
ggplot(data, aes(x = as.factor(Light_Cond_ID))) + 
  geom_bar(fill = "orange") +
  labs(title = "Number of Accidents by Light Condition", 
       x = "Light Condition", 
       y = "Number of Accidents") +
  theme_minimal()

# Create a bar plot showing the number of accidents by speed limit category
ggplot(data, aes(x = as.factor(Crash_Speed_LimitCat))) + 
  geom_bar(fill = "steelblue") +
  labs(title = "Number of Accidents by Speed Limit Category", 
       x = "Speed Limit Category", 
       y = "Number of Accidents") +
  theme_minimal()

# Create a bar plot using the existing age group values
ggplot(data, aes(x = as.factor(Prsn_Age))) + 
  geom_bar(fill = "skyblue") +
  labs(title = "Number of Accidents by Age Group", 
       x = "Age Group", 
       y = "Number of Accidents") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Create a bar plot for the number of accidents by injury severity (Prsn_Injry_Sev_ID)
ggplot(data, aes(x = as.factor(Prsn_Injry_Sev_ID))) + 
  geom_bar(fill = "coral") +
  labs(title = "Number of Accidents by Injury Severity", 
       x = "Injury Severity", 
       y = "Number of Accidents") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Bar plot for number of accidents by gender (GenMale)
ggplot(data, aes(x = as.factor(GenMale))) + 
  geom_bar(fill = "lightblue") +
  labs(title = "Number of Accidents by Gender", 
       x = "Gender (1 = Male, 0 = Female)", 
       y = "Number of Accidents") +
  theme_minimal()

# Bar plot for number of accidents by ethnicity (Prsn_Ethnicity_ID)
ggplot(data, aes(x = as.factor(Prsn_Ethnicity_ID))) + 
  geom_bar(fill = "lightgreen") +
  labs(title = "Number of Accidents by Ethnicity", 
       x = "Ethnicity", 
       y = "Number of Accidents") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Bar plot for number of accidents by intersection relation (Intrsct_Relat_ID)
ggplot(data, aes(x = as.factor(Intrsct_Relat_ID))) + 
  geom_bar(fill = "orange") +
  labs(title = "Number of Accidents by Intersection Relation", 
       x = "Intersection Relation", 
       y = "Number of Accidents") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Bar plot for number of accidents by surface dryness (SurfDry)
ggplot(data, aes(x = as.factor(SurfDry))) + 
  geom_bar(fill = "purple") +
  labs(title = "Number of Accidents by Surface Dryness", 
       x = "Surface Dryness (1 = Dry, 0 = Not Dry)", 
       y = "Number of Accidents") +
  theme_minimal()

# Bar plot for number of accidents by traffic control (Traffic_Cntl_ID)
ggplot(data, aes(x = as.factor(Traffic_Cntl_ID))) + 
  geom_bar(fill = "tomato") +
  labs(title = "Number of Accidents by Traffic Control", 
       x = "Traffic Control", 
       y = "Number of Accidents") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Create a summary table with factors and observations
univariate_summary <- data.frame(
  Factor = c("Weather Condition (Wthr_Cond_ID)", 
             "Light Condition (Light_Cond_ID)", 
             "Speed Limit Category (Crash_Speed_LimitCat)", 
             "Age Group (Prsn_Age)", 
             "Injury Severity (Prsn_Injry_Sev_ID)", 
             "Gender (GenMale)", 
             "Ethnicity (Prsn_Ethnicity_ID)", 
             "Intersection Relation (Intrsct_Relat_ID)", 
             "Surface Dryness (SurfDry)", 
             "Traffic Control (Traffic_Cntl_ID)"),
  Observation = c("Majority of crashes occurred in clear weather",
                  "Most crashes occurred in daylight condition",
                  "Crashes follow a normal distribution by speed limit category",
                  "The middle-age group (22-54 years) is more prone to crashes",
                  "Most crashes resulted in no injuries, only property damage",
                  "Males are more prone to crashes",
                  "White and Hispanic individuals are more represented in crashes",
                  "Majority of crashes occurred at non-intersections",
                  "Most crashes happened on dry surfaces",
                  "Crashes are more frequent where traffic control is present")
)

# Display the summary table
knitr::kable(univariate_summary, col.names = c("Factor", "Observation"), caption = "Summary of Univariate Analysis")

# Load necessary libraries
library(ggplot2)
library(dplyr)

# List of categorical columns to be plotted against Prsn_Injry_Sev_ID
categorical_columns <- c("Wthr_Cond_ID", "Light_Cond_ID", "Road_Type_ID", "Road_Algn_ID", 
                         "Traffic_Cntl_ID", "Harm_Evnt_ID", "Intrsct_Relat_ID", 
                         "FHE_Collsn_ID", "Road_Part_Adj_ID", "Road_Cls_ID", 
                         "Pop_Group_ID", "Crash_Speed_LimitCat", "Veh_Body_Styl_ID", 
                         "Prsn_Ethnicity_ID", "Prsn_Age")

# Loop through each categorical column and create a bar plot with respect to Prsn_Injry_Sev_ID
for (col in categorical_columns) {
  plot <- ggplot(data, aes_string(x = col, fill = "Prsn_Injry_Sev_ID")) +
    geom_bar(position = "dodge") +
    labs(title = paste("Distribution of", col, "by Injury Severity"), 
         x = col, 
         y = "Count", 
         fill = "Injury Severity (KABCO)") +
    theme_minimal() +
    theme(axis.text.x = element_text(angle = 45, hjust = 1))
  
  print(plot)  # Use print() to display the plot inside the loop
}

ggplot(data, aes(x = as.factor(GenMale), fill = as.factor(Prsn_Injry_Sev_ID))) + 
  geom_bar(position = "dodge") +
  labs(title = "Injury Severity by Gender", x = "Gender (1 = Male, 0 = Female)", y = "Count", fill = "Injury Severity") +
  theme_minimal()

# Load necessary libraries
library(DT)
library(htmltools)

# Create a summary table with factors and observations
barplot_summary <- data.frame(
  Factor = c("Wthr_Cond_ID", "Light_Cond_ID", "Road_Type_ID", "Road_Algn_ID", 
             "Traffic_Cntl_ID", "Harm_Evnt_ID", "Intrsct_Relat_ID", 
             "FHE_Collsn_ID", "Road_Part_Adj_ID", "Road_Cls_ID", 
             "Pop_Group_ID", "Crash_Speed_LimitCat",
             "Prsn_Ethnicity_ID", "Prsn_Age"),
  Observation = c("Most severe (KA) injuries occur in clear weather", 
                  "Most severe (KA) injuries happen during daylight", 
                  "Most severe (KA) injuries take place on 2-lane, 2-way roads", 
                  "Most severe (KA) injuries occur on straight roads", 
                  "Most severe (KA) injuries occur around dividers or marked lanes", 
                  "Motor vehicles are most prone to severe (KA) injuries", 
                  "Most severe (KA) injuries happen on non-intersection roads", 
                  "The major cause of KA injuries is rear-end collisions", 
                  "Most severe (KA) injuries occur on proper roads", 
                  "Most severe (KA) injuries happen on state highways", 
                  "Rural populations are the main victims of KA injuries", 
                  "Most severe (KA) injuries occur due to high speeds (65-70 mph)", 
                  "White people are the major victims of crashes", 
                  "Middle-aged individuals are the main victims of KA crashes")
)

# Title for the interactive table
htmltools::tagList(
  tags$h4("Summary Table: Injury Severity by categories"),
  
  # Create an interactive DataTable
  datatable(barplot_summary, 
            options = list(pageLength = 10, autoWidth = TRUE),
            colnames = c("Factor", "Observation"))
)

# Load necessary libraries
library(ggplot2)
library(MASS)  # For calculating contingency tables

# Function to calculate contingency coefficient
contingency_coefficient <- function(table) {
  chi2 <- chisq.test(table)$statistic
  n <- sum(table)  # Total observations
  return(sqrt(chi2 / (chi2 + n)))
}

# List of categorical columns
categorical_columns <- c("Wthr_Cond_ID", "Light_Cond_ID", "Road_Type_ID", "Road_Algn_ID", 
                         "Traffic_Cntl_ID", "Harm_Evnt_ID", "Intrsct_Relat_ID", 
                         "FHE_Collsn_ID", "Road_Part_Adj_ID", "Road_Cls_ID", 
                         "Pop_Group_ID", "Crash_Speed_LimitCat", "Prsn_Ethnicity_ID", "Prsn_Age")

# Calculate the contingency coefficient for each categorical column
contingency_results <- data.frame(Categorical_Variable = character(), Contingency_Coefficient = numeric())

for (col in categorical_columns) {
  # Create a contingency table for each variable with respect to Prsn_Injry_Sev_ID
  contingency_table <- table(data$Prsn_Injry_Sev_ID, data[[col]])
  
  # Calculate the contingency coefficient
  coeff <- contingency_coefficient(contingency_table)
  
  # Append the result to the results dataframe
  contingency_results <- rbind(contingency_results, 
                               data.frame(Categorical_Variable = col, Contingency_Coefficient = coeff))
}

# Plot the results using ggplot2
ggplot(contingency_results, aes(x = reorder(Categorical_Variable, Contingency_Coefficient), y = Contingency_Coefficient)) + 
  geom_bar(stat = "identity", fill = "skyblue") +
  coord_flip() +  # Flip coordinates for better readability
  labs(title = "Contingency Coefficients for Categorical Variables", x = "Categorical Variable", y = "Contingency Coefficient") +
  theme_minimal()

# Function to calculate p-value using Chi-square test
calculate_p_value <- function(x, y) {
  # Create a contingency table
  tbl <- table(x, y)
  
  # Perform chi-square test
  chi2_test <- chisq.test(tbl)
  
  # Return the p-value
  return(chi2_test$p.value)
}

# List of categorical columns
categorical_columns <- c("Wthr_Cond_ID", "Light_Cond_ID", "Road_Type_ID", "Road_Algn_ID", 
                         "Traffic_Cntl_ID", "Harm_Evnt_ID", "Intrsct_Relat_ID", 
                         "FHE_Collsn_ID", "Road_Part_Adj_ID", "Road_Cls_ID", 
                         "Pop_Group_ID", "Crash_Speed_LimitCat", "Prsn_Ethnicity_ID", "Prsn_Age")

# Data frame to store p-values
p_values_results <- data.frame(Categorical_Variable = character(), P_Value = numeric())

# Loop through each categorical variable and calculate the p-value
for (col in categorical_columns) {
  # Calculate the p-value for each variable with respect to Prsn_Injry_Sev_ID
  p_value <- calculate_p_value(data$Prsn_Injry_Sev_ID, data[[col]])
  
  # Append the result to the results dataframe
  p_values_results <- rbind(p_values_results, 
                            data.frame(Categorical_Variable = col, P_Value = p_value))
}

# Show the results
print(p_values_results)

# Optionally, filter the results for variables with p-values less than 0.05 (significant associations)
significant_results <- p_values_results %>% filter(P_Value < 0.05)
print("Significant Results (p < 0.05):")

print(significant_results)

# Visualize the p-values using ggplot2
ggplot(p_values_results, aes(x = reorder(Categorical_Variable, P_Value), y = P_Value)) + 
  geom_bar(stat = "identity", fill = "skyblue") +
  coord_flip() +  # Flip coordinates for better readability
  labs(title = "P-Value Analysis for Categorical Variables", x = "Categorical Variable", y = "P-Value") +
  theme_minimal() +
  geom_hline(yintercept = 0.05, color = "red", linetype = "dashed")  # Add a threshold line at p = 0.05

significant_results <- p_values_results %>% filter(P_Value < 0.05)
print("Significant Results (p < 0.05):")

print(significant_results)

# Enhanced Violin Plot
ggplot(data, aes(x = as.factor(Prsn_Injry_Sev_ID), y = TrafVol)) + 
  geom_violin(fill = "lightblue", color = "black", alpha = 0.7, trim = FALSE) +  # Violin plot with some transparency and no trimming
  geom_boxplot(width = 0.1, color = "black", outlier.shape = NA, alpha = 0.5) +  # Boxplot inside the violin for better clarity
  geom_jitter(width = 0.1, size = 1, alpha = 0.3, color = "#34495e") +  # Add jittered points to show individual data points
  labs(title = "Traffic Volume Distribution by Injury Severity", 
       x = "Injury Severity (KABCO scale)", 
       y = "Traffic Volume") +
  theme_minimal() +  # Use minimal theme for cleaner look
  theme(
    axis.text.x = element_text(angle = 45, hjust = 1),  # Rotate x-axis labels for better readability
    plot.title = element_text(hjust = 0.5)  # Center the plot title
  ) +
  scale_y_continuous(labels = scales::comma)  # Use comma format for y-axis to make large numbers easier to read

# Boxplot for Traffic Volume (TrafVol) by Injury Severity (Prsn_Injry_Sev_ID)
ggplot(data, aes(x = as.factor(Prsn_Injry_Sev_ID), y = TrafVol)) + 
  geom_boxplot(fill = "lightblue") +
  labs(title = "Boxplot of Traffic Volume by Injury Severity", 
       x = "Injury Severity (KABCO scale)", 
       y = "Traffic Volume") +
  theme_minimal()

numeric_cols <- data[, sapply(data, is.numeric)]
cor_matrix <- cor(numeric_cols, use = "complete.obs")
print(cor_matrix)

library(corrplot)
corrplot(cor_matrix, method = "circle", type = "lower")

# Create a contingency table (Injury Severity Vs Speed Limit)
contingency_table_speed <- table(data$Crash_Speed_LimitCat, data$Prsn_Injry_Sev_ID)

# Convert to dataframe for ggplot
contingency_df_speed <- as.data.frame(contingency_table_speed)

# Heatmap for Injury Severity vs Speed Limit Category
ggplot(contingency_df_speed, aes(Var1, Var2, fill = Freq)) +
  geom_tile(color = "white") +
  scale_fill_gradient(low = "lightblue", high = "red") +
  labs(title = "Heatmap of Injury Severity Vs Speed Limit Category", 
       x = "Speed Limit Category", 
       y = "Injury Severity (KABCO scale)", 
       fill = "Frequency") +
  theme_minimal(base_size = 15) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Function to detect outliers using IQR
detect_outliers_iqr <- function(x) {
  Q1 <- quantile(x, 0.25, na.rm = TRUE)
  Q3 <- quantile(x, 0.75, na.rm = TRUE)
  IQR <- Q3 - Q1
  lower_bound <- Q1 - 1.5 * IQR
  upper_bound <- Q3 + 1.5 * IQR
  outliers <- which(x < lower_bound | x > upper_bound)
  return(outliers)
}

# Apply the function to TrafVol (Traffic Volume)
outliers_trafvol <- detect_outliers_iqr(data$TrafVol)

# Create a new column to identify outliers
data$Outlier <- "No"  # Default value
data$Outlier[outliers_trafvol] <- "Yes"  # Mark outliers as "Yes"

# Visualize Outliers with Boxplot and highlight the outliers
ggplot(data, aes(x = "", y = TrafVol)) +
  geom_boxplot(fill = "lightblue") +
  geom_jitter(aes(color = Outlier), width = 0.1, size = 2) +  # Add jittered points and color by outlier status
  scale_color_manual(values = c("No" = "black", "Yes" = "red")) +  # Red for outliers, black for normal points
  labs(title = "Outlier Detection for Traffic Volume", 
       x = "", 
       y = "Traffic Volume") +
  theme_minimal()