EcoSta 2019: Start Registration
View Submission - EcoSta2019
Title: Impact analysis and spatial R$^2$ for spatial autoregressive models: Application to air pollution in China Authors:  Hsuan-Yu Chang - Peking University (China) [presenting]
Jihai Yu - Peking University (China)
Abstract: Impact analysis and its asymptotic inference for spatial autoregressive models are investigated. A spatial version of coefficient of determination (R$^2$) to measure the model fit is also proposed. We first study the cross-section case, where various impacts are introduced to measure the interaction and feedback effects in a space dimension. We then study the spatial dynamic panel case with simultaneous spatial and dynamic feedback involved in the impacts. A R$^2$ is developed for spatial autoregressive models, which is usually not defined for a linear regression model with endogenous regressors. Monte Carlo results show that the impact analysis and spatial R$^2$ has satisfactory finite sample properties. Finally, we apply the impact analysis and the spatial R$^2$ to investigate how the meteorological factors and air pollutants affect PM2.5 in Chinese cities.