BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20261030T213854EDT-9949vFcUkS@132.216.98.100
DTSTAMP:20261031T013854Z
DESCRIPTION:Audrey Béliveau\, PhD\n\nAssistant Professor\n	Department of Sta
 tistics and Actuarial Science |\n	University of Waterloo\n\nWHEN: Wednesday
 \, September 18\, 2024\, from 3:30 to 4:30 p.m.\n	WHERE: Hybrid | 2001 McGi
 ll College Avenue\, Room 1201\; Zoom\n	NOTE: Audrey Béliveau will be presen
 ting in-person\n\nAbstract\n\nNetwork meta-analysis (NMA) is a valuable st
 atistical tool for combining evidence on the comparative efficacy and safe
 ty of medical treatments from multiple studies. In this work\, we develop 
 a penalization framework for NMA that penalizes all pairwise differences b
 etween treatments using a generalized fused lasso (GFL). This approach imp
 roves model parsimony\, resulting in more precise estimates of treatment d
 ifferences and increased statistical power. Practical advantages include: 
 1) no prior knowledge of the similarity of treatment effects is required\,
  2) treatments are only assigned separate ranks if there is sufficient evi
 dence to suggest they are different\, and 3) computing time is minimal. Th
 e novel GFL-NMA method is successfully applied to three separate real-worl
 d NMAs on diabetes\, Parkinson’s disease\, and depression\, where the best
 -fitting GFL-NMA model outperformed the standard NMA model (ΔAICc > 6.5).
 \n\nSpeaker bio\n\nDr. Béliveau is an Associate Professor in the Statistic
 s and Actuarial Science Department at the University of Waterloo. Her rese
 arch interests include network meta-analysis\, Bayesian modeling\, capture
 -recapture methods\, and survey sampling. Please visit: https://uwaterloo.
 ca/scholar/a2belive.\n
DTSTART:20240918T193000Z
DTEND:20240918T203000Z
SUMMARY:A Penalization Method for Improving the Parsimony of Network Meta-A
 nalysis Models
URL:https://www.mcgill.ca/epi-biostat-occh/channels/event/penalization-meth
 od-improving-parsimony-network-meta-analysis-models-359518
END:VEVENT
END:VCALENDAR