Olympic Medal Quantity Forecasting: a Random Forest Algorithm-based Model Construction
Year: 2025
Type: article
Abstract: Against the backdrop of the unstoppable wave of globalization in sports, the competition for Olympic medals has shown an increasingly fierce trend. Countries have invested a lot of resources to improve their performance in the Olympic Games in order to be in a favorable position in the medal competition. In this study, a random forest model is developed to predict the number of gold medals and the total number of medals of each country in the 2028 Olympic Games. Firstly, the data were obtained from the official website of the Olympic Games and data preprocessing was carried out. After completing data cleaning and organizing, a series of key influence indicators such as whether it is the host country, the number of athletes, the total score and so on are introduced, and then a random forest model is built to predict the total number of medals and gold medals of each country. Finally, based on the prediction results, it was determined that in the 2028 Olympic Games, countries such as Cuba, Germany and Slovakia have the potential to achieve breakthroughs, while countries such as Belgium, Ecuador and Israel may experience a decline in the acquisition of medals. This study breaks through the limitations of linear assumptions in traditional econometric models, utilizes the nonlinear fitting ability of the Random Forest algorithm to capture complex variable interactions, and quantifies the dynamic impact of the 'host effect' on the distribution of medals, and reveals the role weights of the core factors such as historical performance and participation size through characteristic contribution analysis. Meanwhile, the prediction results can provide scientific basis for the National Olympic Committees to optimize resource allocation and formulate strategies, sports economic research and event public opinion prediction.
DOI: 10.61784/wjit3038
Authors: JunBo Zhu, LinFeng Li
Institution: Chongqing University of Science and Technology
Language: English
Subfield: Economics and Econometrics
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4411459256