Abstract

We investigate the properties of the linear Two-way Fixed Effects (FE) estimator for panel data when the underlying Data Generating Process (DGP) of all observed variables is left completely unspecified. The FE estimator is consistent for some pseudo-true value and characterize the corresponding asymptotic distribution. We show that the rate of convergence is determined by the degree of model misspecification, and that the asymptotic distribution can be non-normal. Furthermore, we argue that uniform non-conservative inference is impossible in this setup, and propose a novel Autoregressive Double Adaptive Wild (AdaWild) bootstrap procedure applicable for a large class of DGPs. Our bootstrap procedure is easy to compute, and Monte Carlo simulations show that it performs well. We use data from U.S. manufacturing industries to illustrate the benefits of our procedure.

Artūras Juodis is is an Assistant Professor at the Amsterdam School of Economics with the focus on econometric theory of panel data models. His current research on econometrics of misspecified panel data models is financed by the NWO Veni grant. His research was published in the Journal of Econometrics, the Journal of Business and Economic Statistics, and the Journal of Applied Econometrics, among others.