Non-parametric Methods for Doubly Robust Estimation of Continuous Treatment Effects

Edward H. Kennedy1, Zongming Ma1, Matthew D. McHugh1, Dylan S. Small1
1University of Pennsylvania, Philadelphia, USA

Tóm tắt

SummaryContinuous treatments (e.g. doses) arise often in practice, but many available causal effect estimators are limited by either requiring parametric models for the effect curve, or by not allowing doubly robust covariate adjustment. We develop a novel kernel smoothing approach that requires only mild smoothness assumptions on the effect curve and still allows for misspecification of either the treatment density or outcome regression. We derive asymptotic properties and give a procedure for data-driven bandwidth selection. The methods are illustrated via simulation and in a study of the effect of nurse staffing on hospital readmissions penalties.

Từ khóa


Tài liệu tham khảo

Andrews, 1994, Handbook of Econometrics, 2247, 10.1016/S1573-4412(05)80006-6

Bickel, 1993, Efficient and Adaptive Estimation for Semiparametric Models

Díaz, 2012, Population intervention causal effects based on stochastic interventions, Biometrics, 68, 541, 10.1111/j.1541-0420.2011.01685.x

Díaz, 2013, Targeted data adaptive estimation of the causal dose-response curve, J. Causl Inf., 1, 171, 10.1515/jci-2012-0005

Fan, 1996, Local Polynomial Modelling and Its Applications

Galvao, 2015, Uniformly semiparametric efficient estimation of treatment effects with a continuous treatment, J. Am. Statist. Ass., 110, 1528, 10.1080/01621459.2014.978005

Gill, 2001, Causal inference for complex longitudinal data: the continuous case, Ann. Statist., 29, 1785, 10.1214/aos/1015345962

Györfi, 2002, A Distribution-free Theory of Nonparametric Regression, 10.1007/b97848

Härdle, 1988, How far are automatically chosen regression smoothing parameters from their optimum?, J. Am. Statist. Ass., 83, 86

Hill, 2011, Bayesian nonparametric modeling for causal inference, J. Computnl Graph. Statist., 20, 217, 10.1198/jcgs.2010.08162

Hirano, 2004, Applied Bayesian Modeling and Causal Inference from Incomplete-data Perspectives, 73, 10.1002/0470090456.ch7

Horowitz, 2009, Semiparametric and Nonparametric Methods in Econometrics, 10.1007/978-0-387-92870-8

Imai, 2004, Causal inference with general treatment regimes, J. Am. Statist. Ass., 99, 854, 10.1198/016214504000001187

Imbens, 2004, Nonparametric estimation of average treatment effects under exogeneity: a review, Rev. Econ. Statist., 86, 4, 10.1162/003465304323023651

Kang, 2007, Demystifying double robustness: a comparison of alternative strategies for estimating a population mean from incomplete data, Statist. Sci., 22, 523

van der Laan, 2003, Unified cross-validation methodology for selection among estimators and a general cross-validated adaptive epsilon-net estimator: finite sample oracle inequalities and examples

van der Laan, 2007, Super Learner, Statist. Appl. Genet. Molec. Biol., 6, 10.2202/1544-6115.1309

van der Laan, 1998, Locally efficient estimation with current status data and time-dependent covariates, J. Am. Statist. Ass., 93, 693, 10.1080/01621459.1998.10473721

van der Laan, 2003, Unified Methods for Censored Longitudinal Data and Causality, 10.1007/978-0-387-21700-0

van der Laan, 2011, Targeted Learning: Causal Inference for Observational and Experimental Data, 10.1007/978-1-4419-9782-1

van der Laan, 2001, Comments on inference for semiparametric models: some questions and an answer, Statist. Sin., 11, 910

Li, 2004, Cross-validated local linear nonparametric regression, Statist. Sin., 14, 485

Li, 2007, Nonparametric Econometrics: Theory and Practice

McHugh, 2013, Hospitals with higher nurse staffing had lower odds of readmissions penalties than hospitals with lower staffing, Hlth Aff., 32, 1740, 10.1377/hlthaff.2013.0613

Neugebauer, 2007, Nonparametric causal effects based on marginal structural models, J. Statist. Planng Inf., 137, 419, 10.1016/j.jspi.2005.12.008

Robins, 2000, Statistical Models in Epidemiology, the Environment, and Clinical Trials, 95, 10.1007/978-1-4612-1284-3_2

Robins, 2001, Comments on inference for semiparametric models: some questions and an answer, Statist. Sin., 11, 920

Rubin, 1974, Estimating causal effects of treatments in randomized and nonrandomized studies, J. Educ. Psychol., 66, 688, 10.1037/h0037350

Rubin, 2005, A general imputation methodology for nonparametric regression with censored data

Rubin, 2006, Doubly robust censoring unbiased transformations

Rubin, 2006, Extending marginal structural models through local, penalized, and additive learning

Tsiatis, 2006, Semiparametric Theory and Missing Data

van der Vaart, 2000, Asymptotic Statistics

van der Vaart, 2014, Higher order tangent spaces and influence functions, Statist. Sci., 29, 679, 10.1214/14-STS478

van der Vaart, 2006, Estimating a survival distribution with current status data and high-dimensional covariates, Int. J. Biostatist., 2, 1, 10.2202/1557-4679.1014

van der Vaart, 1996, Weak Convergence and Empirical Processes, 10.1007/978-1-4757-2545-2

Wang, 2010, Nonparametric regression with missing outcomes using weighted kernel estimating equations, J. Am. Statist. Ass., 105, 1135, 10.1198/jasa.2010.tm08463

Wasserman, 2006, All of Nonparametric Statistics