We propose a multiply robust estimator for the Average Treatment Effect Among the Treated (ATT). The proposed estimation procedure can simultaneously accommodate multiple working models for both the propensity score and the conditional mean of the counterfactual outcome given covariates. In addition, it can explicitly balance a set of user-specified moments of the covariate distributions between the treatment groups. The resulting estimator is consistent if any working model is correctly specified. With the data generating process typically unknown for observational studies, the proposed method provides substantial robustness against possible model misspecifications compared to existing estimators of the ATT. Simulation results show the excellent finite sample performance of the proposed estimator.

References

• Abadie, A., & Imbens, G. W. (2006). Large sample properties of matching estimators for average treatment effects. Econometrica, 74(1), 235–267. https://doi.org/10.1111/ecta.2006.74.issue-1
• Bang, H., & Robins, J. M. (2005). Doubly robust estimation in missing data and causal inference models. Biometrics, 61(4), 962–973. https://doi.org/10.1111/biom.2005.61.issue-4
• Cao, W., Tsiatis, A. A., & Davidian, M. (2009). Improving efficiency and robustness of the doubly robust estimator for a population mean with incomplete data. Biometrika, 96, 723–734. https://doi.org/10.1093/biomet/asp033
• Chan, K. C. G., & Yam, S. C. P. (2014). Oracle, multiple robust and multipurpose calibration in a missing response problem. Statistical Science, 29(3), 380–396.
• Chan, K. C. G., Yam, S. C. P., & Zhang, Z. (2016). Globally efficient non-parametric inference of average treatment effects by empirical balancing calibration weighting. Journal of the Royal Statistical Society Series B: Statistical Methodology, 78(3), 673–700. https://doi.org/10.1111/rssb.12129
• Chen, J., Sitter, R. R., & Wu, C. (2002). Using empirical likelihood methods to obtain range restricted weights in regression estimators for surveys. Biometrika, 89(1), 230–237. https://doi.org/10.1093/biomet/89.1.230
• Chen, S., & Haziza, D. (2017). Multiply robust imputation procedures for the treatment of item nonresponse in surveys. Biometrika, 104(2), 439–453.
• Deville, J., & Särndal, C.-E. (1992). Calibration estimators in survey sampling. Journal of the American Statistical Association, 87(418), 376–382. https://doi.org/10.1080/01621459.1992.10475217
• Duan, X., & Yin, G. (2017). Ensemble approaches to estimating the population mean with missing response. Scandinavian Journal of Statistics, 44(4), 899–917. https://doi.org/10.1111/sjos.v44.4
• Fan, J., Imai, K., Lee, I., Liu, H., Ning, Y., & Yang, X. (2023). Optimal covariate balancing conditions in propensity score estimation. Journal of Business and Economic Statistics, 41(1), 97–110. https://doi.org/10.1080/07350015.2021.2002159
• Hainmueller, J. (2012). Entropy balancing for causal effects: A multivariate reweighting method to produce balanced samples in observational studies. Political Analysis, 20(1), 25–46. https://doi.org/10.1093/pan/mpr025
• Han, P. (2012). A note on improving the efficiency of inverse probability weighted estimator using the augmentation term. Statistics and Probability Letters, 82(12), 2221–2228. https://doi.org/10.1016/j.spl.2012.08.005
• Han, P. (2014a). A further study of the multiply robust estimator in missing data analysis. Journal of Statistical Planning and Inference, 148, 101–110. https://doi.org/10.1016/j.jspi.2013.12.006
• Han, P. (2014b). Multiply robust estimation in regression analysis with missing data. Journal of the American Statistical Association, 109(507), 1159–1173. https://doi.org/10.1080/01621459.2014.880058
• Han, P. (2016a). Combining inverse probability weighting and multiple imputation to improve robustness of estimation. Scandinavian Journal of Statistics, 43(1), 246–260. https://doi.org/10.1111/sjos.v43.1
• Han, P. (2016b). Intrinsic efficiency and multiple robustness in longitudinal studies with drop-out. Biometrika, 103(3), 683–700. https://doi.org/10.1093/biomet/asw024
• Han, P., Kong, L., Zhao, J., & Zhou, X. (2019). A general framework for quantile estimation with incomplete data. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 81(2), 305–333. https://doi.org/10.1111/rssb.12309
• Han, P., & Wang, L. (2013). Estimation with missing data: Beyond double robustness. Biometrika, 100(2), 417–430. https://doi.org/10.1093/biomet/ass087
• Hernán, M. A., & Robins, J. M. (2018). Causal inference. Chapman & Hall/CRC.
• Hirano, K., Imbens, G. W., & Ridder, G. (2003). Efficient estimation of average treatment effects using the estimated propensity score. Econometrica, 71(4), 1161–1189. https://doi.org/10.1111/ecta.2003.71.issue-4
• Imbens, G. W., & Rubin, D. B. (2015). Causal inference for statistics, social, and biomedical sciences: An introduction. Cambridge University Press.
• Kang, J. D. Y., & Schafer, J. L. (2007). Demystifying double robustness a comparison of alternative strategies for estimating a population mean from incomplete data. Statistical Science, 22(4), 523–539.
• Kim, J. K. (2010). Calibration estimation using exponential tilting in sample surveys. Survey Methodology, 36(2), 145–155.
• Kim, J. K., & Park, M. (2010). Calibration estimation in survey sampling. International Statistical Review, 78(1), 21–39. https://doi.org/10.1111/insr.2010.78.issue-1
• Li, W., Yang, S., & Han, P. (2020). Robust estimation for moment condition models with data missing not at random. Journal of Statistical Planning and Inference, 207, 246–254. https://doi.org/10.1016/j.jspi.2020.01.001
• Molina, J., Rotnitzky, A., Sued, M., & Robins, J. (2017). Multiple robustness in factorized likelihood models. Biometrika, 104, 561–581. https://doi.org/10.1093/biomet/asx027
• Newey, W. K., & Smith, R. J. (2004). Higher order properties of GMM and generalized empirical likelihood estimators. Econometrica, 72(1), 219–255. https://doi.org/10.1111/ecta.2004.72.issue-1
• Qin, J., & Lawless, J. (1994). Empirical likelihood and general estimating equations. The Annals of Statistics, 22(1), 300–325. https://doi.org/10.1214/aos/1176325370
• Qin, J., Shao, J., & Zhang, B. (2008). Efficient and doubly robust imputation for covariate-dependent missing responses. Journal of the American Statistical Association, 103(482), 797–810. https://doi.org/10.1198/016214508000000238
• Qin, J., & Zhang, B. (2007). Empirical-likelihood-based inference in missing response problems and its application in observational studies. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 69(1), 101–122. https://doi.org/10.1111/j.1467-9868.2007.00579.x
• Robins, J. M., Rotnitzky, A., & Zhao, L. P. (1994). Estimation of regression coefficients when some regressors are not always observed. Journal of the American Statistical Association, 89(427), 846–866. https://doi.org/10.1080/01621459.1994.10476818
• Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41–55. https://doi.org/10.1093/biomet/70.1.41
• Rosenbaum, P. R., & Rubin, D. B. (1985). Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. The American Statistician, 39(1), 33–38.
• Rotnitzky, A., Lei, Q., Sued, M., & Robins, J. M. (2012). Improved double-robust estimation in missing data and causal inference models.. Biometrika, 99, 439–456. https://doi.org/10.1093/biomet/ass013
• Schennach, S. (2007). Point estimation with exponentially tilted empirical likelihood. Annals of Statistics, 35(2), 634–672. https://doi.org/10.1214/009053606000001208
• Shi, X., Miao, W., Nelson, J. C., & Tchetgen Tchetgen, E. (2020). Multiply robust causal inference with double-negative control adjustment for categorical unmeasured confounding. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 82(2), 521–540. https://doi.org/10.1111/rssb.12361
• Tan, Z. (2010). Bounded, efficient and doubly robust estimation withinverse weighting. Biometrika, 97(3), 661–682. https://doi.org/10.1093/biomet/asq035
• van der Laan, M. J., & Gruber, S. (2010). Collaborative double robust targeted maximum likelihood estimation. The International Journal of Biostatistics, 6(1), Article 17.
• van der Vaart, A. W. (1998). Asymptotic statistics. Cambridge University Press.
• Wang, L. (2019). Multiple robustness estimation in causal inference. Communications in Statistics - Theory and Methods, 48(23), 5701–5718. https://doi.org/10.1080/03610926.2018.1520881
• Wang, L., & Tchetgen Tchetgen, E. (2018). Bounded, efficient and multiply robust estimation of average treatment effects using instrumental variables. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 80(3), 531–550. https://doi.org/10.1111/rssb.12262
• Wu, C., & Sitter, R. R. (2001). A model-calibration approach to using complete auxiliary information from survey data. Journal of the American Statistical Association, 96(453), 185–193. https://doi.org/10.1198/016214501750333054
• Zhang, S., Han, P., & Wu, C. (2022). Calibration techniques encompassing survey sampling, missing data analysis and causal inference. International Statistical Review. https://doi.org/10.1111/insr.12518 .
• Zhao, Q., & Percival, D. (2017). Entropy balancing is doubly robust. Journal of Causal Inference, 5(1). https://doi.org/10.1515/jci-2016-0010

To cite this article: Lu Wang & Peisong Han (2024) Multiply robust estimation for average treatment effect among treated, Statistical Theory and Related Fields, 8:1, 29-39, DOI: 10.1080/24754269.2023.2293554

To link to this article: https://doi.org/10.1080/24754269.2023.2293554