In this study, multiple linear regression (MLR) method was used to establish the rela-tionship between the O3 at time t + 1 and other indices including hourly air pollutant concentra-tions and meteorological conditions at time t. Then O3 was predicted using the obtained best-fitting MLR. The results indicated that the relationship between the O3 at time t + 1 and other indices including hourly air pollutant concentrations and meteorological conditions at time t agreed with MLR well, The values of mean absolute percentage error (MAPE), correlation co-
efficient (R), coefficient of determination (R2), mean square error (MSE), and root mean square error (RMSE) were 29.09 %, 0.95, 0.90, 45.33, and 6.73, respectively when determining the best-fitting equation. In addition, MLR could predict hourly ozone concentrations successfully.
The values of MAPE, R, R2, MSE, and RMSE were 10.37 %, 0.93, 0.86, 0.33, and 0.57, respec-tively when predicting. It also indicated that the hourly air pollutant concentrations and mete-orological conditions at time t could be applied on the prediction of ozone of time t +1.