Prediction and Analysis of Olympic Medals Based on Probit Regression and Tobit Model

This paper aims to predict and analyze the distribution of Olympic medals by constructing a two-stage prediction framework based on Probit regression and the Tobit model. First, we preprocess the historical Olympic data, including data cleaning, integration, and visualization analysis, to reveal the medal performance characteristics of different countries across various historical periods. On this basis, we use the Probit regression model to conduct binary classification predictions on whether countries will win medals and apply the Tobit model to predict the number of medals for the winning countries. The results show that the model can effectively predict the medal performance of countries in the Olympics, with high goodness-of-fit and prediction accuracy. Additionally, the robustness of the model is verified through sensitivity analysis.