Sage Journals: Discover world-class research

Abstract

Inverse probability weighting can be used to estimate the average treatment effect in propensity score analysis. When there is lack of overlap in the propensity score distributions between the treatment groups under comparison, some weights may be excessively large, causing numerical instability and bias in point and variance estimation. We study a class of modified inverse probability weighting estimators that can be used to avoid this problem. These weights cause the estimand to deviate from the average treatment effect. We provide some justification for this deviation from the perspective of treatment effect discovery. We show that when lack of overlap occurs, the modified weights can achieve substantial gains in statistical power compared with inverse probability weighting and other propensity score methods. We develop analytical variance estimates that properly adjust for the sampling variability of the estimated propensity scores, and augment the modified inverse probability weighting estimator with outcome models for improved efficiency, a property that resembles double robustness. Results from extensive simulations and a real data application support our conclusions. The proposed methodology is implemented in R package PSW.

Keywords

Average treatment effect doubly robust estimation observational study propensity score weighting statistical power

Get full access to this article

View all access options for this article.

References

Rosenbaum

Rubin

. The central role of the propensity score in observational studies for causal effects. Biometrika 1983; 70: 41–55.

Hirano

Imbens

. Estimation of causal effects using propensity score weighting: An application to data on right heart catheterization. Health Serv Outcomes Res Methodol 2001; 2: 259–278.

Hirano

Imbens

Ridder

. Efficient estimation of average treatment effects using the estimated propensity score. Econometrica 2003; 71: 1161–1189.

Lunceford

Davidian

. Stratification and weighting via the propensity score in estimation of causal treatment effects: a comparative study. Stat Med 2004; 23: 2937–2960.

Rubin

. Bayesian inference for causal effects: the role of randomization. Ann Stat 1978; 6: 34–58.

Busso

DiNardo

McCrary

. New evidence on the finite sample properties of propensity score reweighting and matching estimators. Rev Econ Stat 2014; 96: 885–897.

Kang

Schafer

. Demystifying double robustness: a comparison of alternative strategies for estimating a population mean from incomplete data. Stat Sci 2007; 22: 523–539.

Lee

Lessler

Stuart

. Weight trimming and propensity score weighting. PLoS One 2011; 6: e18174–e18174.

Crump

Hotz

Imbens

, et al. Dealing with limited overlap in estimation of average treatment effects. Biometrika 2009; 96: 187–199.

10.

Yang S and Ding P. Asymptotic inference of causal effects with observational studies trimmed by the estimated propensity scores. Biometrika 2018; 105: 487–493.

11.

Greene

. A weighting analogue to pair matching in propensity score analysis. Int J Biostat 2013; 9: 215–234.

12.

Crump

Hotz

Imbens

, et al. Moving the goalposts: addressing limited overlap in the estimation of average treatment effects by changing the estimand. NBER Technical Rep 2006; 330: 1–48.

13.

Morgan

Zaslavsky

. Balancing covariates via propensity score weighting. J Am Stat Assoc 2016; 113: 390–400.

14.

Zou

Shuster

, et al. On variance estimate for covariate adjustment by propensity score analysis. Stat Med 2016; 35: 3537–3548.

15.

Abadie

Imbens

. Matching on the estimated propensity score. Econometrica 2016; 84: 781–807.

16.

Williamson

Morley

Lucas

, et al. Variance estimation for stratified propensity score estimators. Stat Med 2012; 31: 1617–1632.

17.

Rubin

. Which ifs have causal answers. J Am Stat Assoc 1986; 81: 961–962.

18.

Imbens

. Nonparametric estimation of average treatment effects under exogeneity: a review. Rev Econ Stat 2004; 86: 4–29.

19.

Rassen

Shelat

Franklin

, et al. Matching by propensity score in cohort studies with three treatment groups. Epidemiology 2013; 24: 401–409.

20.

Stürmer

Rothman

Avorn

, et al. Treatment effects in the presence of unmeasured confounding: dealing with observations in the tails of the propensity score distribution – a simulation study. Am J Epidemiol 2010; 172: 843–854.

21.

Imbens

Wooldridge

. Recent developments in the econometrics of program evaluation. J Econ Lit 2009; 47: 5–86.

22.

Hernán

Alonso

Logan

, et al. Observational studies analyzed like randomized experiments: an application to postmenopausal hormone therapy and coronary heart disease. Epidemiology 2008; 19: 766–779.

23.

Hernán

Robins

. Authors’ response, part i: observational studies analyzed like randomized experiments: best of both worlds. Epidemiology 2008; 19: 789–792.

24.

Robins

Hernan

Brumback

. Marginal structural models and causal inference in epidemiology. Epidemiology 2000; 11: 550–560.

25.

Bang

Robins

. Doubly robust estimation in missing data and causal inference models. Biometrics 2005; 61: 962–973.

26.

Freedman

Berk

. Weighting regressions by propensity scores. Eval Rev 2008; 32: 392–409.

27.

Austin

. Type I error rates, coverage of confidence intervals, and variance estimation in propensity-score matched analyses. Int J Biostat 2009; 5: 13–13.

28.

Imai

King

, et al. Matchit: nonparametric preprocessing for parametric causal inference. J Stat Softw 2011; 42: 1–28.

29.

Hill

Reiter

. Interval estimation for treatment effects using propensity score matching. Stat Med 2006; 25: 2230–2256.

30.

Gutman

Rubin

. Estimation of causal effects of binary treatments in unconfounded studies with one continuous covariate. Stat Meth Med Res 2017; 26: 1199–1215.

31.

Schafer

Kang

. Average causal effects from nonrandomized studies: a practical guide and simulated example. Psychol Meth 2008; 13: 279–279.

32.

Abadie

Imbens

. On the failure of the bootstrap for matching estimators. Econometrica 2008; 76: 1537–1557.

33.

Austin

Small

. The use of bootstrapping when using propensity-score matching without replacement: a simulation study. Stat Med 2014; 33: 4306–4319.

34.

Abadie

Imbens

. Large sample properties of matching estimators for average treatment effects. Econometrica 2006; 74: 235–267.

35.

Austin

. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009; 28: 3083–3107.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.16 MB

Propensity score weighting analysis and treatment effect discovery

Abstract

Keywords

Get full access to this article

References

Supplementary Material