Selection correction in panel data models: An application to the estimation of females' wage equations

Abstract

In recent years a number of panel estimators have been suggested for sample selection models, where both the selection equation and the equation of interest contain individual effects which are correlated with the explanatory variables. Not many studies exist that use these methods in practise. We present and compare alternative estimators, and apply them to a typical problem in applied econometrics: the estimation of the wage returns to experience for females. We discuss the assumptions each estimator imposes on the data, and the problems that occur in our applications. This should be particularly useful to practitioners who consider using such estimators in their own application. All estimators rely on the assumption of strict exogeneity of regressors in the equation of interest, conditional on individual specific effects and the selection mechanism. This assumption is likely to be violated in many applications. Also, life history variables are often measured with error in survey data sets, because they contain a retrospective component. We show how this particular measurement error, and not strict exogeneity can be taken into account within the estimation methods discussed.