Prediction of Olympic Medal Counts Based on Multilevel Negative Binomial Regression Model and Bayesian Methods

Abstract

The aim of this study is to accurately predict the number of medals for each country in the 2028 Summer Olympics in Los Angeles by constructing a multilevel negative binomial regression model combined with a Bayesian approach. This study focuses on predicting the medal count for each country in the 2028 Summer Olympics in Los Angeles by constructing a multilevel negative binomial regression model. The model integrates factors such as historical medal data, economic and demographic indicators, and the host country effect. A Bayesian approach, specifically the Markov Chain Monte Carlo (MCMC) method, is employed to estimate the parameters, allowing for the quantification of uncertainty and capturing heterogeneity across countries and Olympic sessions. The model not only predicts the number of medals but also provides prediction intervals to highlight countries with likely improvements or declines in their performance. This approach provides more accurate and reliable predictions compared to previous studies, offering insights for the Olympic Committee to optimize resource allocation and strategy planning for participating countries. The model's validity is confirmed through performance indicators such as mean square error and coefficient of determination, ensuring its accuracy and reliability for future predictions. By considering the complex interplay of various factors, this model offers a comprehensive framework for understanding the dynamics of Olympic medal distribution, which can be further refined and adapted for subsequent Olympic Games and other international sporting events.