Simula@BI: This Shock is Different: Estimation and Inference in Misspecied Two-Way Fixed Effects Panel Regressions?
Speaker: Arturas Juodis
We would like to welcome you to Simula@bi’s next research seminar in data science. The seminar series consists of talks on fundamental and applied research within statistics, machine learning and artificial intelligence.
This week’s speaker is Arturas Juodis from Amsterdam School of Economics, University of Amsterdam and Tinbergen Institute. The title of the talk is "This Shock is Different: Estimation and Inference in Misspeci
ed Two-Way Fixed Effects Panel Regressions?".
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.