Entropy balancing is doubly robust

Abstract

Covariate balance is a conventional key diagnostic for methods used estimating causal effects from observational studies. Recently, there is an emerging interest in directly incorporating covariate balance in the estimation. We study a recently proposed entropy maximization method called Entropy Balancing (EB), which exactly matches the covariate moments for the different experimental groups in its optimization problem. We show EB is doubly robust with respect to linear outcome regression and logistic propensity score regression, and it reaches the asymptotic semiparametric variance bound when both regressions are correctly specified. This is surprising to us because there is no attempt to model the outcome or the treatment assignment in the original proposal of EB. Our theoretical results and simulations suggest that EB is a very appealing alternative to the conventional weighting estimators that estimate the propensity score by maximum likelihood.

23 pages, 6 figures, Journal of Causal Inference 2016

References in corpus (5)

• Comment: Performance of Double-Robust Estimators When ``Inverse Probability'' Weights Are Highly Variable
• Comment: Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data
• Comment: Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data
• Rejoinder: Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data
• Covariate Balancing Propensity Score by Tailored Loss Functions

Cited by in corpus (25)

• Minimal Dispersion Approximately Balancing Weights: Asymptotic Properties and Practical Considerations
• Improving trial generalizability using observational studies
• Transporting Experimental Results with Entropy Balancing
• Generalized Optimal Matching Methods for Causal Inference
• Covariate Balancing Propensity Score by Tailored Loss Functions
• Estimating causal effects with optimization-based methods: A review and empirical comparison
• A Framework for Covariate Balance using Bregman Distances
• Entropy Balancing for Causal Generalization with Target Sample Summary Information
• Varying impacts of letters of recommendation on college admissions: Approximate balancing weights for subgroup effects in observational studies
• Sparsity Double Robust Inference of Average Treatment Effects
• Robust Estimation of Propensity Score Weights via Subclassification
• The Statistical Performance of Matching-Adjusted Indirect Comparisons
• Large Sample Properties of Matching for Balance
• Kpop: A kernel balancing approach for reducing specification assumptions in survey weighting
• Reducing bias in difference-in-differences models using entropy balancing
• Double Robust Representation Learning for Counterfactual Prediction
• Treatment Effects Estimation by Uniform Transformer
• Deconfounding Scores: Feature Representations for Causal Effect Estimation with Weak Overlap
• A Balancing Weight Framework for Estimating the Causal Effect of General Treatments
• End-to-End Balancing for Causal Continuous Treatment-Effect Estimation
• How balance and sample size impact bias in the estimation of causal treatment effects: A simulation study
• Optimal Estimation of Generalized Average Treatment Effects using Kernel Optimal Matching
• Kernel-Distance-Based Covariate Balancing
• A Python Library For Empirical Calibration
• On matching-adjusted indirect comparison and calibration estimation