BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20250809T121300EDT-7955K2imGE@132.216.98.100 DTSTAMP:20250809T161300Z DESCRIPTION:Penalized doubly-robust estimation of adaptive treatment strate gies\n\nZeyu Bian\, University of Miami\n Tuesday January 17\, 12-1pm\n Zoom Link: https://mcgill.zoom.us/j/86855481591\n\nAbstract: Adaptive treatmen t strategies (ATSs) are often estimated from data sources with many covari ates measured\, only a subset of which are useful for tailoring treatment or control of confounding. In such cases\, including all the covariates in the analytic model could possibly yield an inappropriate or needlessly co mplicated treatment decision. Hence\, it is crucial to apply variable sele ction techniques to ATSs. Variable selection with the objective of optimiz ing treatment decisions has been the subject of only very little literatur e. In this talk\, I will present a regression-based estimation method that can naturally incorporate variable selection through a penalization appro ach that incorporates sparsity while ensuring strong heredity\, and show h ow we can additionally incorporate confounder selection into the approach. We illustrate the methods using data from a pilot sequential multiple ass ignment randomized trial of a web-based\, stress management intervention u sing a stepped-care method for cardiovascular diseases patients to determi ne useful tailoring variables while adjusting for chance imbalances in imp ortant covariates due to the smaller sample size in the pilot (joint work with Zeyu Bian\, Sahir Bhatnagar\, and Susan Shortreed)\n DTSTART:20220117T170000Z DTEND:20220117T180000Z SUMMARY:QLS Seminar - Zeyu Bian URL:/qls/channels/event/qls-seminar-zeyu-bian-344128 END:VEVENT END:VCALENDAR