An Introduction to Proximal Causal Inference
Eric J. Tchetgen Tchetgen, PhD
Professor of Statistics and Data Science | Wharton School |
University of Pennsylvania
Where: Virtual | Zoom
A standard assumption for causal inference from observational data is that one has measured a sufficiently rich set of covariates to ensure that within covariates strata, subjects are exchangeable across observed treatment values. Skepticism about the exchangeability assumption in observational studies is often warranted because it hinges on one's ability to accurately measure covariates capturing all potential sources of confounding. Realistically, confounding mechanisms can rarely if ever, be learned with certainty from measured covariates. One can therefore only ever hope that covariate measurements are at best proxies of true underlying confounding mechanisms operating in an observational study, thus invalidating causal claims made on basis of standard exchangeability conditions. Causal learning from proxies is a challenging inverse problem which has to date remained unresolved. In this paper, we introduce a formal potential outcome framework for proximal causal learning, which while explicitly acknowledging covariate measurements as imperfect proxies of confounding mechanisms, offers an opportunity to learn about causal effects in settings where exchangeability on basis of measured covariates fails. Sufficient conditions for nonparametric identification are given, leading to the proximal g-formula and corresponding proximal g-computation algorithm for estimation, both generalizations of Robins' foundational g-formula and g-computation algorithm, which account explicitly for bias due to unmeasured confounding. Both point treatment and time-varying treatment settings are considered, and an application of proximal g-computation of causal effects is given for illustration.
- Reasoning about unmeasured confounding using proxies
- Nonparametric identification
Eric J Tchetgen Tchetgen is The Luddy Family President's Distinguished Professor and Professor of Statistics and Data Science at the Wharton School of the University of Pennsylvania. He also co-directs the Penn Center for Causal Inference, which supports the development and dissemination of causal inference methods in Health and Social Sciences. He has published extensively on Causal Inference, Missing Data and Semiparametric Theory with several impactful applications ranging from HIV research, Genetic Epidemiology, Environmental Health and Alzheimer's Disease and related aging disorders. He is an Amazon scholar working with Amazon scientists on a variety of causal inference problems in the Tech industry space. Professor Tchetgen Tchetgen is an 2022 inaugural co-recipient of the newly established Rousseeuw Prize for statistics in recognition for his work in Causal Inference with applications in Public Health and Medicine.