BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:Statistics Seminar Counterfactual Q-learning via the linear Bu
 ckley– James method for longitudinal survival data
X-WR-TIMEZONE:Central Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260808T072853Z
UID:tag:localist.com\,2008:EventInstance_51065059452932
DTSTART:20251024T160000Z
DTEND:20251024T170000Z
DESCRIPTION:Professor Jong-Min Kim\nUniversity of Minnesota-Morris\n\n \n\n
 Abstract\n\nTreatment strategies are critical in healthcare\, particularly
  when outcomes are subject to censoring. This study introduces the Counter
 factual Buckley–James Q-Learning framework\, which integrates counterfac
 tual reasoning with the Buckley–James method and reinforcement learning 
 to address challenges arising from longitudinal survival data. The Buckley
 –James method imputes censored survival times via conditional expectatio
 ns based on observed data\, offering a robust mechanism for handling incom
 plete outcomes. By incorporating these imputed values into a counterfactua
 l Q-learning framework\, the proposed method enables the estimation and co
 mparison of potential outcomes under different treatment strategies. This 
 facilitates the identification of optimal dynamic treatment regimes that m
 aximize expected survival time. Through extensive simulation studies\, the
  method demonstrates robust performance across various sample sizes and ce
 nsoring scenarios\, including right censoring and missing at random. Appli
 cation to real- world clinical trial data further highlights the utility o
 f this approach in informing personalized treatment decisions\, providing 
 an interpretable and reliable tool for optimizing survival outcomes in com
 plex clinical settings.
LOCATION:Hume Hall\, 331
SUMMARY:Statistics Seminar Counterfactual Q-learning via the linear Buckley
 – James method for longitudinal survival data
URL;VALUE=URI:https://eventcalendar.olemiss.edu/event/statistics-seminar-co
 unterfactual-q-learning-via-the-linear-buckley-james-method-for-longitudin
 al-survival-data
CATEGORIES:Lectures
END:VEVENT
END:VCALENDAR
