Semiparametric two-stage estimation of sample selection models subject to Tobit-type selection rules

Abstract

A semiparametric two-stage estimation method is proposed for the estimation of sample selection models which are subject to Tobit-type selection rules. With randomization restrictions on the disturbances of the model, all the regression coefficients in the model are, in general, identifiable without exclusion restrictions. The proposed estimator is shown to be [radical sign]n-consistent and asymptotically normal. Some Monte Carlo results, to demonstrate its finite sample performance, are provided.