Consistent Estimation of Panel Data Models with a Multi-factor Error Structure

This paper considers the panel data model with a multifactor structure in both the errors and the regressors which was studied by Pesaran (“Estimation and Inference in Large Heterogeneous Panels with a Multifactor Error Structure, Econometrica, 2006). Estimators are proposed that are consistent for fixed T as N tends to infinity. By allowing T to be fixed some of the assumptions imposed by Pesaran are relaxed and, at the same time, some of the complexities of the large N and T asymptotics are bypassed. A small Monte Carlo simulation shows that these new estimators are very accurate for very small values of T .