Study on Medal Prediction Based on XGBoost and ARIMA Models
Jiayi Ma, Yiyuan Chen, Xiang Lin · Scientific Journal of Technology · 2025
Focusing on predictive modelling research, this paper constructs an Olympic medal prediction model and evaluates its performance. XGBoost combined with Monte Carlo simulation is used to predict the number of Olympic medals. Optuna optimises the hyperparameters and evaluates the model in terms of mean squared error (MSE), root mean squared error (RMSE) and mean absolute error (MAE), and the results show that the model performs well in predicting the number of gold medals and the total number of medals; Monte Carlo simulation is used to determine the prediction intervals and to evaluate the model uncertainty. For the first medal prediction, the ARIMA model was used to process the data with time series characteristics, and the ADF test was used to judge the stability of the data, construct and solve the model to predict the trend of the number of gold medals; at the same time, the logistic regression model was used to convert the medal count labels into binary labels for training, and ROCAUC was used to assess the classification performance of the model, and the code outputs its value as 0.7403825829634327 , indicating that the model has the ability to distinguish whether awards.