Minimizing Bias in Selection on Observables Estimators When Unconfoundness Fails

38 Pages Posted: 18 Apr 2008 Last revised: 12 May 2008

See all articles by Daniel L. Millimet

Daniel L. Millimet

Southern Methodist University (SMU) - Department of Economics; IZA Institute of Labor Economics

Rusty Tchernis

Georgia State University - Department of Economics; National Bureau of Economic Research (NBER); IZA Institute of Labor Economics

Multiple version iconThere are 2 versions of this paper

Date Written: April 16, 2008

Abstract

We characterize the bias of propensity score based estimators of common average treatment effect parameters in the case of selection on unobservables. We then propose a new minimum biased estimator of the average treatment effect. We assess the finite sample performance of our estimator using simulated data, as well as a timely application examining the causal effect of the School Breakfast Program on childhood obesity. We find our new estimator to be quite advantageous in many situations, even when selection is only on observables.

Keywords: Treatment Effects, Propensity Score, Bias, Unconfoundedness, Selection on Unobservables

JEL Classification: C21, C52

Suggested Citation

Millimet, Daniel L. and Tchernis, Rusty, Minimizing Bias in Selection on Observables Estimators When Unconfoundness Fails (April 16, 2008). CAEPR Working Paper No. 2008-008, Available at SSRN: https://ssrn.com/abstract=1121765 or http://dx.doi.org/10.2139/ssrn.1121765

Daniel L. Millimet (Contact Author)

Southern Methodist University (SMU) - Department of Economics ( email )

Box 750496
Dallas, TX 75275
United States
214-768-3269 (Phone)
214-768-1821 (Fax)

HOME PAGE: http:\\www.smu.edu\~millimet

IZA Institute of Labor Economics

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

Rusty Tchernis

Georgia State University - Department of Economics ( email )

P.O. Box 3992
Atlanta, GA 30302-3992
United States

HOME PAGE: http://www2.gsu.edu/~ecort

National Bureau of Economic Research (NBER) ( email )

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

IZA Institute of Labor Economics ( email )

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