Quarterly PM2.5 prediction using a novel seasonal grey model and its further application in health effects and economic loss assessment: evidences from Shanghai and Tianjin, China

Abstract

Previous research only focused on PM2.5 prediction without considering its further application or just evaluated the past years' health effects and economic losses caused by PM2.5 without studying the future scenarios. Thus, a novel hybrid system using a seasonal grey model with the fractional order accumulation, called SFGM (1, 1), and health economic loss assessment model was developed in this study, which can not only perform quarterly PM2.5 prediction, but also estimate its health effects and economic losses. The results indicated that (1) the designed SFGM (1, 1) can not only reflect the seasonal fluctuation, but also predict the seasonal PM2.5 concentrations with higher prediction accuracy in both out-and-in-samples than comparison models. (2) The total economic losses in 2020 of Shanghai and Tianjin will be 6867.25 million yuan (95% CI: 3072.34–10704.47) and 4869.20 million yuan (95% CI: 2194.50–7532.00), respectively, showing that Shanghai will suffer bigger economic losses than Tianjin. (3) The economic loss caused by the premature deaths attributable to PM2.5 is the largest, accounting for more than 70% of the total economic loss. Finally, the findings can help policymakers to formulate more policies and take effective measures to improve public awareness of environmental protection.