AQI Prediction Model of Shijiazhuang City Based on Multivariate Time Series Analysis

With the rapid development of industrial industry, smoke, automobile exhaust and industrial exhaust emissions, various air pollution problems have become increasingly prominent. In order to prevent and control urban air pollution problems, it is necessary to understand the causes of air pollution and accurately grasp the changing trend of air quality (AQI). However, when traditional multiple linear regression and time series models predict AQI, there are some problems such as incomplete information utilization and low prediction accuracy. Based on this, the paper puts forward an AQI prediction method based on multiple time series (ARMAX). In the method, the lag order of AQI influencing factors in time series is introduced into the regression term, the residual series information is extracted, and the AQI prediction model is established. Taking Shijiazhuang city as an example, the data of AQI and its influencing factors from October 2013 to May 2022 in Shijiazhuang were collected. After preprocessing the data, the multiple linear regression model, the single time series model and the ARMAX model were used to fit and predict the data. After that, RMSE, MAE and MAPE are used to compare the prediction accuracy of the three models. The results show that the prediction accuracy of ARMAX model is significantly higher than that of single time series model or multiple linear regression model. Finally, according to the experimental conclusion, the article gives some suggestions for the prevention and control of air pollution in Shijiazhuang.