CMStatistics 2022: Start Registration
View Submission - CMStatistics
B0562
Title: One-step weighting for the generalization of causal inference Authors:  Ambarish Chattopadhyay - Stanford University (United States) [presenting]
Eric Cohn - Harvard University (United States)
Jose Zubizarreta - Harvard University (United States)
Abstract: Weighting methods are often used to generalize estimates of causal effects from a study sample to a target population. Traditional methods construct the weights by separately modeling the treatment assignment and the study selection probabilities and then multiplying functions (e.g., inverses) of the estimated probabilities. These estimated multiplicative weights may not produce adequate covariate balance and can be highly variable, resulting in biased and/or unstable estimators, particularly when there is limited covariate overlap across populations or treatment groups. To address these limitations, we propose a weighting approach for both randomized and observational studies that weights each treatment group directly in `one go' towards the target population. We present a general framework for generalization problems by characterizing the study and target populations in terms of probability distributions. Under this framework, we justify this one-step weighting approach. By construction, this approach directly balances covariates relative to the target population and produces stable weights. Moreover, this approach does not require individual-level data from the target population. We connect this approach to inverse probability and inverse odds weighting. We show that the one-step weighting estimator for the target average treatment effect is consistent, asymptotically Normal, doubly robust, and semiparametrically efficient.