A Comparative Study on Predictive Performance of Some Advanced Count Regression Models and Machine Learning Models

Author

Binta Saji John

Overview

This repository contains the R Markdown files and analysis conducted as part of my M.Sc. Statistics dissertation.

Dissertation Title

A Comparative Study on Predictive Performance of Some Advanced Count Regression Models and Machine Learning Models

Contents

R Markdown Files

• Overdispersion Analysis
• Underdispersion Analysis
• Zero-Inflated Count Data Analysis

Methodology

The study evaluates advanced count regression models and machine learning techniques using simulated and real count data.

Models Considered

• Poisson Regression
• Negative Binomial Regression
• Zero-Inflated Models
• Hurdle Models
• Random Forest
• XGBoost
• LightGBM
• GLMNET
• Neural Networks

Software

• R
• RStudio
• R Markdown

Research Area

• Count Data Modelling
• Statistical Learning
• Machine Learning
• Biostatistics
• Predictive Analytics