BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.4.0.1//EN TZID:Europe/Paris X-WR-TIMEZONE:Europe/Paris BEGIN:VEVENT UID:411@isdm.umontpellier.fr DTSTART;TZID=Europe/Paris:20240917T140000 DTEND;TZID=Europe/Paris:20240917T140000 DTSTAMP:20260828T120049Z URL:https://isdm.umontpellier.fr/events/data-integration-approaches-to-est imate-heterogeneous-treatment-effects-2/ SUMMARY:Data integration approaches to estimate heterogeneous treatment eff ects DESCRIPTION:Online\n\nMachine Learning in Montpellier\, Theory &\; Pract ice\n\nIn many fields\, including medicine\, education\, and public policy \, researchers and practitioners are interested in determining what works for whom - in other words\, identifying subgroups for whom specific interv entions work particularly well. By finding these subgroups\, we can more e ffectively focus resources and give interventions to those who might benef it the most. These individualized treatment decisions can improve outcomes \, but answering these types of questions is challenging with a single dat aset. Specifically\, randomized controlled trials have unconfounded treatm ent assignment but are often too small to reliably estimate heterogeneous treatment effects\, while larger observational datasets might suffer from confounding. Data integration methods can utilize the benefits of differen t sources of data while accounting for bias. In this talk\, I first discus s non-parametric data integration approaches for combining multiple random ized controlled trials to estimate the effect of treatments conditional on observed characteristics. I explore the performance of these methods thro ugh a simulation study\, and I apply the approaches to four randomized con trolled trials to examine effect heterogeneity of treatments for major dep ressive disorder. I then discuss methods for applying these multi study tr eatment effect models to an external\, observational target sample represe nted by electronic health records of a set of patients. With these methods \, we can utilize individual-level data across sources to improve our abil ity to make intervention decisions that are tailored to individuals or com munities\, and we can ultimately apply our conclusions to a given target p opulation.\n\nMachine Learning in Montpellier\, Theory &\; Practice ATTACH;FMTTYPE=image/jpeg:https://isdm.umontpellier.fr/wp-content/uploads/ 2026/06/ml-mtp-gC78d5.png CATEGORIES:ML MTP END:VEVENT BEGIN:VTIMEZONE TZID:Europe/Paris X-LIC-LOCATION:Europe/Paris BEGIN:DAYLIGHT DTSTART:20240331T030000 TZOFFSETFROM:+0100 TZOFFSETTO:+0200 TZNAME:CEST END:DAYLIGHT END:VTIMEZONE END:VCALENDAR