BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Instrumental Variable Approaches To Individualized Treatment Regim
 es Under A Counterfactual World
DTSTART:20220429T140000
DTEND:20220429T150000
DTSTAMP:20260407T045627Z
UID:3b243cbb59e0ac5f3f799b5e37a68bb57ee3af76a36fd1f4908fd14a
CATEGORIES:Conferences - Seminars
DESCRIPTION:Yifan Cui\, National University of Singapore\nThere is a fast-
 growing literature on estimating optimal treatment regimes based on random
 ized trials or observational studies under a key identifying condition of 
 no unmeasured confounding. Because confounding by unmeasured factors canno
 t generally be ruled out with certainty in observational studies or random
 ized trials subject to noncompliance\, we propose a robust classification-
 based instrumental variable approach to learning optimal treatment regimes
  under endogeneity. Specifically\, we establish identification of both val
 ue function for a given regime and optimal regimes with the aid of a binar
 y instrumental variable\, when no unmeasured confounding fails to hold.\n\
 nWe also construct novel multiply robust classification-based estimators. 
 In addition\, we propose to identify and estimate optimal treatment regime
 s among those who would comply to the assigned treatment under a monotonic
 ity assumption. Furthermore\, we consider the problem of individualized tr
 eatment regimes under sign and partial identification. In the former case\
 , i) we provide a necessary and sufficient identification condition of opt
 imal treatment regimes with an instrumental variable\; ii) we establish th
 e somewhat surprising result that complier optimal regimes can be consiste
 ntly estimated without directly collecting compliance information and ther
 efore without the complier average treatment effect itself being identifie
 d. In the latter case\, we establish a formal link between individualized 
 decision making under partial identification and classical decision theory
  under uncertainty through a unified lower bound perspective.\n\n\n 
LOCATION:GR A3 30 https://plan.epfl.ch/?room==GR%20A3%2030 https://epfl.zo
 om.us/j/66136073806
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR