# JACKKNIFE AND BOOTSTRAP

# Load necessary packages
library(readr)
library(dplyr)

# Function to perform bootstrap analysis
bootstrap_analysis <- function(data, num_samples = 10000, confidence_level = 0.95) {
  n <- length(data)
  means <- numeric(num_samples)
  
  for (i in 1:num_samples) {
    sample <- sample(data, size = n, replace = TRUE)
    means[i] <- mean(sample)
  }
  
  mean_val <- mean(means)
  bias <- mean_val - mean(data)
  std_error <- sd(means)
  ci_lower <- quantile(means, (1 - confidence_level) / 2)
  ci_upper <- quantile(means, (1 + confidence_level) / 2)
  
  list(mean = mean_val, bias = bias, std_error = std_error, ci_lower = ci_lower, ci_upper = ci_upper)
}

# Function to perform Jackknife analysis
jackknife_analysis <- function(data) {
  n <- length(data)
  jackknife_means <- numeric(n)
  
  for (i in 1:n) {
    jackknife_sample <- data[-i]
    jackknife_means[i] <- mean(jackknife_sample)
  }
  
  jackknife_mean <- mean(jackknife_means)
  jackknife_bias <- (n - 1) * (jackknife_mean - mean(data))
  jackknife_std_error <- sqrt((n - 1) * mean((jackknife_means - jackknife_mean)^2))
  
  list(mean = jackknife_mean, bias = jackknife_bias, std_error = jackknife_std_error)
}

# Prompt user to select a CSV file (HF.csv <- available in the Supplementary Materials)
cat("Please select the CSV file:\n")
file_path <- file.choose()

# Load the CSV file
df <- read_csv(file_path)

# List available headers in the file
cat("Available headers in the CSV file:\n")
headers <- colnames(df)
for (i in seq_along(headers)) {
  cat(paste0(i - 1, ": ", headers[i], "\n"))
}

repeat {
  # Prompt user to select a header
  selected_header <- readline("\nEnter the number of the header you want to evaluate (or press 'X' to finish): ")
  
  if (tolower(selected_header) == 'x') {
    cat("Exiting the analysis.\n")
    break
  }
  
  header_index <- as.integer(selected_header)
  
  if (is.na(header_index) || header_index < 0 || header_index >= length(headers)) {
    cat("Invalid number. Please select a valid number from the list.\n")
    next
  }
  
  # Select the header and remove missing values
  selected_column <- headers[header_index + 1]
  data <- df[[selected_column]] %>% na.omit()
  
  if (length(data) == 0) {
    cat("The selected column contains no valid data after removing missing values. Please choose another column.\n")
    next
  }
  
  # Perform bootstrap analysis
  bootstrap_results <- bootstrap_analysis(data)
  
  # Perform Jackknife analysis
  jackknife_results <- jackknife_analysis(data)
  
  # Display Bootstrap results
  cat("\nBootstrap Analysis for '", selected_column, "':\n", sep = "")
  cat("Mean: ", round(bootstrap_results$mean, 2), "\n")
  cat("Bias: ", round(bootstrap_results$bias, 3), "\n")
  cat("Standard Error: ", round(bootstrap_results$std_error, 3), "\n")
  cat("95% Confidence Interval: (", round(bootstrap_results$ci_lower, 2), ", ", round(bootstrap_results$ci_upper, 2), ")\n\n", sep = "")
  
  # Display Jackknife results
  cat("Jackknife Analysis for '", selected_column, "':\n", sep = "")
  cat("Jackknife Mean: ", round(jackknife_results$mean, 2), "\n")
  cat("Jackknife Bias: ", round(jackknife_results$bias, 3), "\n")
  cat("Jackknife Standard Error: ", round(jackknife_results$std_error, 3), "\n\n", sep = "")
}