BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20251220T191341EST-4872urIJx5@132.216.98.100
DTSTAMP:20251221T001341Z
DESCRIPTION:Abstract:  Bayesian statistical approaches to mixed treatment c
 omparisons (MTCs) are becoming more popular due to their flexibility and i
 nterpretability. Many randomized clinical trials report multiple outcomes 
 with possible inherent correlations. Moreover\, MTC data are typically spa
 rse and researchers often choose study arms based on the outcomes of previ
 ous trials. In this talk\, we summarize existing hierarchical Bayesian met
 hods for MTCs with a single outcome\, and subsequently introduce novel Bay
 esian approaches for multiple outcomes simultaneously\, rather than in sep
 arate MTC analyses. We do this by incorporating missing data and correlati
 on structure between outcomes through contrast- and arm-based parameteriza
 tions that consider any unobserved treatment arms as missing data to be im
 puted. We also extend the model to apply to all types of generalized linea
 r model outcomes\, such as count or continuous responses. We develop a new
  measure of inconsistency under our missing data framework\, having more s
 traightforward interpretation and implementation than standard methods. We
  off  er a simulation study under various missingness mechanisms (e.g.\, M
 CAR\, MAR\, and MNAR) providing evidence that our models outperform existi
 ng models in terms of bias and MSE\, then illustrate our methods with two 
 real MTC datasets. We close with a discussion of our results and a few ave
 nues for future methodological development. This talk represents joint wor
 k with Hwanhee Hong and Haitao Chu of the University of Minnesota.  BIO:  
 Brad Carlin is Mayo Professor in Public Health and Professor and Head of t
 he  Division of Biostatistics at the University of Minnesota. He has publi
 shed more than 135 papers in refereed books and journals\, and supervised 
 or co-supervised 19 PhD dissertations to completion. He is co-author of th
 ree popular textbooks\, “Bayesian Methods for Data Analysis” with Tom Loui
 s\, “Hierarchical Modeling and Analysis for Spatial Data” with Sudipto Ban
 erjee and Alan Gelfand\, and 'Bayesian Adaptive Methods for Clinical Trial
 s' with Scott Berry\, J. Jack Lee\, and Peter Muller.  He is a winner of t
 he Mortimer Spiegelman Award from the APHA\, and from 2006-2009 served as 
 editor-in-chief of Bayesian Analysis\, the official journal of the Interna
 tional Society for Bayesian Analysis (ISBA).  Prof. Carlin has extensive e
 xperience teaching short courses and tutorials\, and has won teaching awar
 ds from both the faculty and the graduate students at the University of Mi
 nnesota\, as well as from the Joint Statistical Meetings CE program. Durin
 g his spare time\, Brad is a musician and bandleader\, providing keyboards
  and vocals in a variety of venues\, some of the more interesting of which
  are visible by typing the phrase 'Bayesian cabaret' into the search windo
 w at YouTube. 
DTSTART:20130226T090000Z
DTEND:20130226T100000Z
LOCATION:Room 25\, Purvis Hall\, CA\, QC\, Montreal\, H3A 1A2\, 1020 avenue
  des Pins Ouest
SUMMARY:Biostatistics Seminar - Dr. Bradley P. Carlin - A Bayesian Missing 
 Data Framework for Multiple Outcome Mixed Treatment Comparisons.
URL:/epi-biostat-occh/channels/event/biostatistics-sem
 inar-dr-bradley-p-carlin-bayesian-missing-data-framework-multiple-ou
END:VEVENT
END:VCALENDAR