摘要
Abstract
This study employed an interval spatial autoregressive panel model to analyze the spatial correlation of PM 2.5 across 27 cities in the Yangtze River Delta,as well as the impacts of PM10,SO2,CO,NO2,and O3 on PM2.5.By comparing four models—spatial autoregression models based on the centre and range method,spatial autoregression models based on the min-max method,spatial autoregression models based on the centre method,and spatial autoregression models based on the parametrized method,we obtained spatial correlation coefficients and impact coefficients while evaluating the fitting accuracy of the models.Results indicated that the spatial autoregression models based on the centre and range method exhibited the best fitting performance.PM2.5 in the Yangtze River Delta urban agglomeration demonstrated a positive spatial spillover effect.PM10 and CO positively influenced PM2.5 concentrations,whereas O3 exerted a negative impact,and the effects of SO2 and NO2 were statistically insignificant.Using the spatial autoregression models based on the centre and range method,this study further forecasted the mass concentration of PM 2.5 in Hangzhou,Hefei,Nanjing,and Shanghai.The results showed a high degree of fit between the predicted values and the actual values.关键词
PM2.5/区间型数据/空间自回归面板模型/区间预测Key words
PM2.5/interval data/spatial autoregressive panel model/interval prediction分类
数理科学
