NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

On the Failure of the Bootstrap for Matching Estimators

Alberto Abadie, Guido W. Imbens

NBER Technical Working Paper No. 325
Issued in June 2006
NBER Program(s):   TWP

Matching estimators are widely used for the evaluation of programs or treatments. Often researchers use bootstrapping methods for inference. However, no formal justification for the use of the bootstrap has been provided. Here we show that the bootstrap is in general not valid, even in the simple case with a single continuous covariate when the estimator is root-N consistent and asymptotically normally distributed with zero asymptotic bias. Due to the extreme non-smoothness of nearest neighbor matching, the standard conditions for the bootstrap are not satisfied, leading the bootstrap variance to diverge from the actual variance. Simulations confirm the difference between actual and nominal coverage rates for bootstrap confidence intervals predicted by the theoretical calculations. To our knowledge, this is the first example of a root-N consistent and asymptotically normal estimator for which the bootstrap fails to work.

download in pdf format
   (336 K)

email paper

This paper is available as PDF (336 K) or via email.

Machine-readable bibliographic record - MARC, RIS, BibTeX

Published: -Abadie, Alberto and Guido W. Imbens. "On the Failure of the Bootstrap for Matching Estimators." Econometrica 76, 6 (2008): 1537-1557.

Users who downloaded this paper also downloaded these:
Abadie and Imbens w15301 Matching on the Estimated Propensity Score
Abadie and Imbens t0283 Simple and Bias-Corrected Matching Estimators for Average Treatment Effects
Abadie and Imbens w14756 A Martingale Representation for Matching Estimators
Heckman, Ichimura, Smith, and Todd w6699 Characterizing Selection Bias Using Experimental Data
Imbens t0237 The Role of the Propensity Score in Estimating Dose-Response Functions
 
Publications
Activities
Meetings
Data
People
About

Support
National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email: info@nber.org

Contact Us