The matching methods have been extensively used to evaluate the impact of a specific treatment in the case as instance of non-experimental studies. In this context, the matching methods based on propensity scores have become a well-known standard.
However it appears in many cases that the propensity score matching methods are not really a panacea. Following the work of Sekhon and Grieve we explore by a Monte Carlo study, the genetic matching method in comparison to the traditional methods.
Our results are contrasted and don't proof the superiority of the last matching method.
Lay, P.: Genetic Matching: Eine bessere Alternative zu den bisherigen Matching-Methoden? Eine empirische Studie über die Auswirkung der Projektförderung der Kommission für Technologie und Innovation, Masterarbeit, Freiburg, 23. Juni 2010, pp. 177
Sekhon, J.S. and Grieve, R.: A New Non-Parametric Matching Method for Bias Adjustment With Applications to Economic Evaluations, Working Paper, 2008, pp. 46
Arvanitis, S.; Donzé, L.; Sydow, N.: Impact of Swiss technology policy on firm innovation performance: an evaluation based on a matching approach, Science and Public Policy, 37, 2010, pp. 63-78.
Keywords: Matching Methods; Genetic Matching; Observational Studies; Monte Carlo Results
Biography: Laurent Donzé (1960) has obtained his PHD in Econometrics in 1990 at the University of Fribourg. He has worked from 1990-1996 on different research projects. From 1996-2002, he was a research fellow at the Swiss Federal Institute of Technology in Zurich (KOF ETH Zurich). Since 2002, he is an Associated Professor at the Department of Quantitative Economics, University of Fribourg.
His teaching experiences are: Mathematics and econometrics (1993, 1995) and Introductury Statistics, Applied Econometrics, and Survey statistics since 2002 at the University of Fribourg.
His main fields of research are in Survey methodology and applied statistics.
He his referee for several journals and is Research Professor for the Swiss Federal Institute of Technology in Zurich (KOF ETH Zurich).