Binary choice panel data models with predetermined variables

Abstract

(i) The explanatory variables are predetermined but not strictly exogenous. This includes lagged dependent variables as well as other forms of unspeci.ed feedback. (ii) Individual e1ects are allowed to be correlated with the explanatory variables. Dependence is speci.ed through the conditional expectation of the e1ects which is let to be non-parametric. We also present a GMM estimator for these models, which is consistent and asymptotically normal for .xed T and large N. We study its .nite sample properties in an autoregressive model by means of Monte Carlo simulations. Finally, as an empirical illustration, we estimate a female labour force participation equation withpredetermined children using PSID data.