R Workshop: Statistical Analysis of Music Data

Overview

This project is an R-based statistical analysis of a music dataset, focusing on features such as key, acousticness, and other audio characteristics. The analysis includes descriptive statistics, hypothesis testing, and visualization, and is designed for educational purposes in an R Workshop course.

Project Structure

r_workshop/
├── main.R                 # Main R script with all analysis steps
├── README.md              # This file
├── data/                  # Data files
│   ├── data.csv          # Main dataset (large, not included)
│   ├── s1.csv            # Sample 1 (100 observations)
│   └── s2.csv            # Sample 2 (400 observations)
├── images/                # Generated plots
│   ├── 1.jpg             # ECDF of key with cumulative distribution
│   ├── 2.jpg             # Histogram of acousticness
│   ├── 3.jpg             # Histogram of key with normal curve
│   ├── 4.jpg             # ECDF of acousticness with uniform fit
│   └── 5.jpg             # ECDF comparison with Kolmogorov-Smirnov
├── output/                # Generated output files
│   └── rplot.pdf         # PDF containing R plot outputs
└── docs/                  # Documentation
    └── README.md         # Detailed project documentation

Requirements

• R (version 4.0 or higher recommended)
• R packages:
  • PerformanceAnalytics
  • samplingbook
  • EnvStats

Install required packages in R:

install.packages(c("PerformanceAnalytics", "samplingbook", "EnvStats"))

Usage

1. Ensure all data files are in the data/ directory
2. Run the main script in R:

   source('main.R')

3. The script will:
   • Output summary statistics to the console
   • Generate plots in the images/ directory
   • Create additional output files in the output/ directory

Analysis Steps

• Descriptive Statistics: Distribution, mean, variance, skewness, and kurtosis for the key variable
• Empirical Distribution Function: Plots ECDF for key and acousticness
• Histograms: Visualizes distributions of acousticness and key
• Confidence Intervals: For mean and variance using normal and t-distributions
• Sample Size Estimation: Calculates required sample size for given error margins
• Normality and Goodness-of-Fit Tests: Pearson's chi-squared, Kolmogorov-Smirnov tests
• Hypothesis Testing: Tests for mean and variance, and comparison between samples

Outputs

• Console Output: Summary statistics, confidence intervals, test results
• Images: Statistical plots saved as JPEG files in images/ directory
• PDF Output: Additional R plot outputs in output/ directory