Bootstrap Inference for K-Nearest Neighbour Matching Estimators

25 Pages Posted: 13 Dec 2010

See all articles by Xavier de Luna

Xavier de Luna

University of Umea - Department of Economics

Per Johansson

IFAU - Institute for Labour Market Policy Evaluation; Uppsala University - Department of Economics; IZA Institute of Labor Economics

Sara Sjostedt

University of Umea - Department Of Computing Science

Abstract

Abadie and Imbens (2008, Econometrica) showed that classical bootstrap schemes fail to provide correct inference for K-nearest neighbour (KNN) matching estimators of average causal effects. This is an interesting result showing that bootstrap should not be applied without theoretical justification. In this paper, we present two resampling schemes, which we show provide valid inference for KNN matching estimators. We resample "estimated individual causal effects" (EICE), i.e. the difference in outcome between matched pairs, instead of the original data. Moreover, by taking differences in EICEs ordered with respect to the matching covariate, we obtain a bootstrap scheme valid also with heterogeneous causal effects where mild assumptions on the heterogeneity are imposed. We provide proofs of the validity of the proposed resampling based inferences. A simulation study illustrates finite sample properties.

Keywords: block bootstrap, subsampling, average causal/treatment effect

JEL Classification: C14, C21

Suggested Citation

de Luna, Xavier and Johansson, Per and Sjostedt-de Luna, Sara, Bootstrap Inference for K-Nearest Neighbour Matching Estimators. IZA Discussion Paper No. 5361, Available at SSRN: https://ssrn.com/abstract=1723999

Xavier De Luna (Contact Author)

University of Umea - Department of Economics ( email )

Umeå University
Umea, SE - 90187
Sweden

Per Johansson

IFAU - Institute for Labour Market Policy Evaluation ( email )

Box 513
751 20 Uppsala
Sweden
+ 46 18 471 70 86 (Phone)
+ 46 18 471 70 71 (Fax)

Uppsala University - Department of Economics ( email )

Uppsala, 751 20
Sweden

IZA Institute of Labor Economics

P.O. Box 7240
Bonn, D-53072
Germany

Sara Sjostedt-de Luna

University of Umea - Department Of Computing Science ( email )

S-90187 Umea
Sweden
+46 (0)90 - 786 51 29 (Phone)

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
68
Abstract Views
541
rank
412,913
PlumX Metrics