Mar 28, 2024
Application of Logistic Regression Model in the Prediction of
Air Quality Level in Zibo City
Objectively evaluating urban ambient air quality and analyzing its influencing factors are of great significance for understanding the
current status of air quality and controlling pollution sources. In this paper, logistic regression analysis is carried out for 366 days of air quality data from January to December 2020 in Zibo City, Shandong Province, with air quality class as the categorical variable, and six variables,
PM2.5, PM10, SO2, CO, NO2, O3, are selected as pollution indicators affecting the air quality in Zibo City, and the stepwise regression method
is utilized to establish a model and determine the weights of each pollution indicator. The established model is used to predict the samples of
Zibo City, and the predicted data and actual data are compared to test the fit of the model. The results show that the logistic regression model
fits well, and SO2 is the strongest factor affecting air quality, and the probability of air pollution is 1.035 times higher for every unit it increases and other variables remain unchanged, providing a basis for controlling the emissions of the primary pollution factors.