English

Fixed Effects Binary Choice Models with Three or More Periods

Econometrics 2022-09-30 v4

Abstract

We consider fixed effects binary choice models with a fixed number of periods TT and regressors without a large support. If the time-varying unobserved terms are i.i.d. with known distribution FF, \cite{chamberlain2010} shows that the common slope parameter is point identified if and only if FF is logistic. However, he only considers in his proof T=2T=2. We show that the result does not generalize to T3T\geq 3: the common slope parameter can be identified when FF belongs to a family including the logit distribution. Identification is based on a conditional moment restriction. Under restrictions on the covariates, these moment conditions lead to point identification of relative effects. If T=3T=3 and mild conditions hold, GMM estimators based on these conditional moment restrictions reach the semiparametric efficiency bound. Finally, we illustrate our method by revisiting Brender and Drazen (2008).

Keywords

Cite

@article{arxiv.2009.08108,
  title  = {Fixed Effects Binary Choice Models with Three or More Periods},
  author = {Laurent Davezies and Xavier D'Haultfoeuille and Martin Mugnier},
  journal= {arXiv preprint arXiv:2009.08108},
  year   = {2022}
}

Comments

Compared to v2, we have in particular added an application (Section 4) and we now discuss the case of unbalanced panel data (Section 3.2)

R2 v1 2026-06-23T18:36:22.257Z