BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260203T131954EST-4631Rt8Pb0@132.216.98.100
DTSTAMP:20260203T181954Z
DESCRIPTION:Eric J. Tchetgen Tchetgen\, PhD\n\nProfessor of Statistics and 
 Data Science | Wharton School |\n	University of Pennsylvania\n\nWhere: Virt
 ual | Zoom\n\nAbstract\n\nA standard assumption for causal inference from 
 observational data is that one has measured a sufficiently rich set of cov
 ariates to ensure that within covariates strata\, subjects are exchangeabl
 e across observed treatment values. Skepticism about the exchangeability a
 ssumption in observational studies is often warranted because it hinges on
  one's ability to accurately measure covariates capturing all potential so
 urces of confounding. Realistically\, confounding mechanisms can rarely if
  ever\, be learned with certainty from measured covariates. One can theref
 ore only ever hope that covariate measurements are at best proxies of true
  underlying confounding mechanisms operating in an observational study\, t
 hus invalidating causal claims made on basis of standard exchangeability c
 onditions. Causal learning from proxies is a challenging inverse problem w
 hich has to date remained unresolved. In this paper\, we introduce a forma
 l potential outcome framework for proximal causal learning\, which while e
 xplicitly acknowledging covariate measurements as imperfect proxies of con
 founding mechanisms\, offers an opportunity to learn about causal effects 
 in settings where exchangeability on basis of measured covariates fails. S
 ufficient conditions for nonparametric identification are given\, leading 
 to the proximal g-formula and corresponding proximal g-computation algorit
 hm for estimation\, both generalizations of Robins' foundational g-formula
  and g-computation algorithm\, which account explicitly for bias due to un
 measured confounding. Both point treatment and time-varying treatment sett
 ings are considered\, and an application of proximal g-computation of caus
 al effects is given for illustration.\n\nLearning Objectives\n\n\n	Reasonin
 g about unmeasured confounding using proxies\n	Nonparametric identification
 \n	g-computation\n\n\nSpeaker Bio\n\nEric 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 So
 cial 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 Alzhe
 imer's Disease and related aging disorders. He is an Amazon scholar workin
 g with Amazon scientists on a variety of causal inference problems in the 
 Tech industry space. Professor Tchetgen Tchetgen is an 2022 inaugural co-r
 ecipient of the newly established Rousseeuw Prize for statistics in recogn
 ition for his work in Causal Inference with applications in Public Health 
 and Medicine.\n\n \n\n \n\n \n\n \n\n \n
DTSTART:20230313T200000Z
DTEND:20230313T210000Z
SUMMARY:An Introduction to Proximal Causal Inference
URL:/epi-biostat-occh/channels/event/introduction-prox
 imal-causal-inference-346771
END:VEVENT
END:VCALENDAR