SEMIPARAMETRIC ESTIMATION OF COUNT REGRESSION-MODELS

This paper develops a semiparametric estimation approach for mixed cou nt regression models based on series expansion for the unknown density of the unobserved heterogeneity. We use the generalized Laguerre seri es expansion around a gamma baseline density to model unobserved heter ogeneity in a Poisson mixture model. We establish the consistency of t he estimator and present a computational strategy to implement the pro posed estimation techniques in the standard count model as well as in truncated, censored, and zero-inflated count regression models. Monte Carlo evidence shows that the finite sample behavior of the estimator is quite good. The paper applies the method to a model of individual s hopping behavior. (C) 1999 Elsevier Science S.A. All rights reserved.