Home  >>  Archives  >>  Volume 17 Number 3  >>  st0485

The Stata Journal
Volume 17 Number 3: pp. 517-545

Subscribe to the Stata Journal

Bias corrections for probit and logit models with two-way fixed effects

Mario Cruz-Gonzalez
Department of Economics
Boston University
Boston, MA
Iván Fernández-Val
Department of Economics
Boston University
Boston, MA
Martin Weidner
Department of Economics
University College London
London, UK
Abstract.  In this article, we present the user-written commands probitfe and logitfe, which fit probit and logit panel-data models with individual and time unobserved effects. Fixed-effects panel-data methods that estimate the unobserved effects can be severely biased because of the incidental parameter problem (Neyman and Scott, 1948, Econometrica 16: 1–32). We tackle this problem using the analytical and jackknife bias corrections derived in Fernández-Val and Weidner (2016, Journal of Econometrics 192: 291–312) for panels where the two dimensions (N and T) are moderately large. We illustrate the commands with an empirical application to international trade and a Monte Carlo simulation calibrated to this application.

View all articles by these authors: Mario Cruz-Gonzalez, Iván Fernández-Val, Martin Weidner

View all articles with these keywords: probitfe, logitfe, probit, logit, panel, fixed effects, bias corrections, incidental parameter problem

Download citation: BibTeX  RIS

Download citation and abstract: BibTeX  RIS