A Study on the Prediction of Olympic Medal Distribution Based on Regression Analysis and ARIMA Models

Abstract

This study aims to construct a predictive model to forecast the distribution of medals at the 2028 Summer Olympics in Los Angeles. Using regression analysis and ARIMA models, we addressed the issues of predicting the number of medals and analyzing the trends in the progress or regression of medals won by each country. First, we employed a linear regression model, incorporating historical medal data and the number of athletes, to predict the number of gold medals and total medals for each country, and calculated the 95% confidence interval. Second, we utilized ARIMA models to analyze the historical trends in medal counts for each country. By integrating feature fusion and machine learning classifiers, we predicted which countries might experience improvements or declines in future Olympics. Additionally, we employed a random forest model to predict the probability of countries that have never won medals securing their first medals. The research findings provide strategic support for national Olympic committees and offer new methodological references for Olympic medal predictions.