Multivariate Statistical Analysis of Athletes' Physical Fitness Data and Optimization of Sports Teaching and Training Strategies

Abstract

The development of physical quality is of great significance in enhancing one's physical health, mastering sports skills, and improving sports level. In this study, the main factors affecting athletes' physical fitness were investigated by principal component analysis, a multiple regression model was constructed to realize the prediction of athletes' physical fitness, and the validity of the model was verified by partial least squares. The 17 indicators of thletes' physical fitness test were analyzed by principal component analysis, and the important factors affecting athletes' physical fitness were categorized into physical characteristics, morphological development, physiological unctions and flexibility and endurance. After analyzing the correlation between each factor and verifying that each factor affects athletes' physical fitness, a regression model was constructed to predict athletes' physical fitness. The partial least squares regression equation was used to predict the four quantitative indexes of physical quality, and the scatter plots of the four indexes were basically symmetrically distributed on the diagonal, indicating that the multiple regression model constructed in this paper is able to predict the physical quality of athletes. In order to help athletes achieve higher quality development, this paper adopts the stratified training method to further optimize the training strategy of physical education.