GMM estimation of spatial panel data models with common factors and a general space–time filter

This paper considers a general spatial panel-data model that incorporates high-order spatial correlation, heterogeneity, common factors and serial correlation in the disturbances, and allows the space and time dynamics to be interacted. The issue of identification is studied, and a generalized method of moments (GMM) estimation is proposed. We show that under certain regularity assumptions, the proposed GMM estimator is consistent and asymptotically normal. The best GMM estimator under normality is also derived. Monte Carlo experiments are conducted to study the finite sample performance of the GMM estimation.