A Comparison of Different Estimators for Panel Data Sample Selection Models

In this paper, we perform an extensive Monte Carlo study of the finite sample properties of different estimators for panel data sample selection models. The estimators investigated are various two-step estimators and maximum likelihood estimators with simultaneous equations for the sampleselection process and the equation of interest. The main result of the MonteCarlo study is that the maximum likelihood estimators of random effectsmodels in general perform better than the two-step estimators.