Enhancing Sea Level Pressure Simulation through Mutual Information Analysis of Meteorological Indicators and Regression Modeling

Abstract

This paper focuses on the meteorological data of Qinghai Province in China and studies the simulation effects of regression models on sea level pressure (SLP). To validate the performance of the models, meteorological data from 2022 is used, and the data is divided into seasons: spring, summer, autumn, and winter. The models are trained using SLP data from different seasons to evaluate their simulation effects. The results indicate significant differences in the performance of the models across different seasons, with lower predictive accuracy in spring and winter, while autumn and summer yield relatively higher prediction performance. The model evaluation metrics include the correlation coefficient R, RMSE, MAE, and MAPE,which provide a comprehensive assessment of the model performance.