Doubly Robust Treatment Effect Estimation with Missing Attributes

• Activity Number: 82 - Evaluating Causal Effects Using Incomplete Data with Interference in Public Health Research
• Type: Topic-Contributed
• Date/Time: Monday, August 9, 2021 : 10:00 AM to 11:50 AM
• Sponsor: ENAR
• Author(s): Imke Mayer* and Erik Sverdrup and Tobias Gauss and Jean-Denis Moyer and Stefan Wager and Julie Josse
• Companies: Centre d'Analyse et de Mathématiques Sociales, EHESS and Graduate School of Business, Stanford University and Department of Anesthesia and Intensive Care, Beaujon Hospital, AP-HP and Department of Anesthesia and Intensive Care, Beaujon Hospital, AP-HP and Stanford University and INRIA Sophia-Antipolis
• Keywords: causal inference; potential outcomes; incomplete confounders; observational data; propensity score estimation; public health

Abstract

Missing attributes are ubiquitous in causal inference, as they are in most applied statistical studies. In this work, we consider various sets of assumptions under which causal inference is possible despite missing attributes and discuss corresponding approaches to average treatment effect estimation, including generalized propensity score methods and multiple imputation. Across an extensive simulation study, we show that no single method systematically out-performs others. We find, however, that doubly robust modifications of standard methods for average treatment effect estimation with missing data repeatedly perform better than their non-doubly robust baselines; for example, doubly robust generalized propensity score methods beat inverse-weighting with the generalized propensity score. This finding is reinforced in an analysis of an observational study on the effect on mortality of tranexamic acid administration among patients with traumatic brain injury in the context of critical care management. Here, doubly robust estimators recover confidence intervals that are consistent with evidence from randomized trials, whereas non-doubly robust estimators do not.