Multiply robust estimation for causal inference problems

Research paper by Cian Naik, Emma J. McCoy, Daniel J. Graham

Indexed on: 08 Nov '16Published on: 08 Nov '16Published in: arXiv - Statistics - Methodology


This paper develops a multiply robust (MR) estimator for causal inference problems involving binary or multivalued treatments. We combine a family of propensity score (PS) models and a family of outcome regression (OR) models to achieve an average potential outcomes estimator that is consistent if just one of the PS or OR models in each family is correctly specified. We provide proofs and simulations that demonstrate multiple robustness in the context of causal inference problems.