Semi-Varying Coefficient Fixed-Effect Panel Data Model Explores the Dynamic Relations between PM2.5and the Meteorological Factors

In this paper we propose a semi-varying coefficient fixed effect panel data model to explore the dynamic relations between PM2.5 and the five meteorological factors: the cumulative wind speed, air pressure, dew point, temperature and hourly precipitation for the five cities: Beijing, Chengdu, Guangzhou, Shanghai and Shenyang, where PM2.5 is a key factor to determine visibility and a serious threat to public health. Then, we combine multivariate local linear fitting, transformation technique with profile likelihood, and establish semi-parametric fixed effects estimators for both parameter vector and varying-coefficient function vector. Finally, we exhibit their estimated dynamic relationships for the five cities in 2015. The proposed procedure can also be generalized to panel data analysis in other fields such as economy and finance.