Robust Inference for Misspecified Models Conditional on Covariates

25 Pages Posted: 23 Sep 2011  

Alberto Abadie

Harvard University - Harvard Kennedy School (HKS); National Bureau of Economic Research (NBER)

Guido W. Imbens

Stanford Graduate School of Business

Fanyin Zheng

affiliation not provided to SSRN

Date Written: September 2011

Abstract

Following the work by White (1980ab; 1982) it is common in empirical work in economics to report standard errors that are robust against general misspecification. In a regression setting these standard errors are valid for the parameter that in the population minimizes the squared difference between the conditional expectation and the linear approximation, averaged over the population distribution of the covariates. In nonlinear settings a similar interpretation applies. In this note we discuss an alternative parameter that corresponds to the approximation to the conditional expectation based on minimization of the squared difference averaged over the sample, rather than the population, distribution of a subset of the variables. We argue that in some cases this may be a more interesting parameter. We derive the asymptotic variance for this parameter, generally smaller than the White robust variance, and we propose a consistent estimator for the asymptotic variance.

Suggested Citation

Abadie, Alberto and Imbens, Guido W. and Zheng, Fanyin, Robust Inference for Misspecified Models Conditional on Covariates (September 2011). NBER Working Paper No. w17442. Available at SSRN: https://ssrn.com/abstract=1932573

Alberto Abadie (Contact Author)

Harvard University - Harvard Kennedy School (HKS) ( email )

79 John F. Kennedy Street
Cambridge, MA 02138
United States
617-496-4547 (Phone)
617-495-2575 (Fax)

National Bureau of Economic Research (NBER)

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

Guido W. Imbens

Stanford Graduate School of Business ( email )

655 Knight Way
Stanford, CA 94305-5015
United States

Fanyin Zheng

affiliation not provided to SSRN ( email )

No Address Available

Paper statistics

Downloads
18
Abstract Views
257