MATCHING AS AN ECONOMETRIC EVALUATION ESTIMATOR

This paper develops the method of matching as an econometric evaluatio n estimator. A rigorous distribution theory for kernel-based matching is presented. The method of matching is extended to more general condi tions than the ones assumed in the statistical literature on the topic . We focus on the method of propensity score matching and show that it is not necessarily better, in the sense of reducing the variance of t he resulting estimator, to use the propensity score method even if pro pensity score is known. We extend the statistical literature on the pr opensity score by considering the case when it is estimated both param etrically and nonparametrically. We examine the benefits of separabili ty and exclusion restrictions in improving the efficiency of the estim ator. Our methods also apply to the econometric selection bias estimat or.