Analysis and Prediction on Sports Performance of Primary School Students

The sports performance of primary school students is influenced by factors like body shape, physiology, and physical fitness. Analysis and prediction on sports performance are of vital importance to facilitate the development of reasonable sports courses and help select talents scientifically for the sports department. In this paper, we aim to use indicators in Physical Fitness Test to provide various prediction models with high generalization performance and high accuracy. We implement correlation analysis and factor analysis to find indicators that greatly influence students' sports performance by using Physical Fitness Test records of students in grade six. Based on the analysis results, we establish three types of prediction models, including multiple linear regression models, factor analysis regression models, and support vector regression models. Adjusted R squared and root mean square error are used to evaluate model performance. The experimental result shows the multiple linear regression model has the best performance on the data set we study, and this result is consistent with previous researches.