Inference in Unbalanced Panel Data Models with Interactive Fixed Effects

Abstract: We derive the asymptotic theory of Bai (2009)'s interactive fixed effects estimator in unbalanced panels where the source of attrition is conditionally random. For inference, we propose a method of alternating projections algorithm based on straightforward scalar expressions to compute the residualized variables required for the estimation of the bias terms and the covariance matrix. Simulation experiments confirm our asymptotic results as reliable finite sample approximations. Furthermore, we reassess Acemoglu et al. (2019). Allowing for a more general form of unobserved heterogeneity, we confirm significant effects of democratization on growth.