Outliers in Semi-Parametric Estimation of Treatment Effects

27 Pages Posted: 2 Nov 2017

See all articles by Darwin Ugarte Ontiveros

Darwin Ugarte Ontiveros

Universidad Privada Boliviana (UPB)

Gustavo J. Canavire-Bacarreza

Inter-American Development Bank (IDB); IZA Institute of Labor Economics

Luis Castro Penarrieta

Universidad Privada Boliviana (UPB)

Date Written: October 30, 2017

Abstract

Average treatment effects estimands can present significant bias under the presence of outliers. Moreover, outliers can be particularly hard to detect, creating bias and inconsistency in the semi-parametric ATE estimands. In this paper, we use Monte Carlo simulations to demonstrate that semi-parametric methods, such as matching, are biased in the presence of outliers. Bad and good leverage points outliers are considered. The bias arises because bad leverage points completely change the distribution of the metrics used to define counterfactuals. Whereas good leverage points increase the chance of breaking the common support condition and distort the balance of the covariates and which may push practitioners to misspecify the propensity score. We provide some clues to diagnose the presence of outliers and propose a reweighting estimator that is robust against outliers based on the Stahel-Donoho multivariate estimator of scale and location. An application of this estimator to LaLonde (1986) data allows us to explain the Dehejia and Wahba (2002) and Smith and Todd (2005) debate on the inability of matching estimators to deal with the evaluation problem.

Keywords: Treatment effects, Outliers, Propensity score, Mahalanobis distance

JEL Classification: C21, C14, C52, C13

Suggested Citation

Ugarte Ontiveros, Darwin and Canavire Bacarreza, Gustavo Javier and Castro Penarrieta, Luis, Outliers in Semi-Parametric Estimation of Treatment Effects (October 30, 2017). Center for Research in Economics and Finance (CIEF), Working Papers, No. 17-20. Available at SSRN: https://ssrn.com/abstract=3063316 or http://dx.doi.org/10.2139/ssrn.3063316

Darwin Ugarte Ontiveros (Contact Author)

Universidad Privada Boliviana (UPB) ( email )

Nuestra Señora de La Paz
Bolivia
Bolivia

Gustavo Javier Canavire Bacarreza

Inter-American Development Bank (IDB) ( email )

1300 New York Avenue NW
Washington, DC 20577
United States

IZA Institute of Labor Economics ( email )

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

Luis Castro Penarrieta

Universidad Privada Boliviana (UPB) ( email )

Nuestra Señora de La Paz
Bolivia
Bolivia

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
14
Abstract Views
224
PlumX Metrics