BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260710T182840EDT-3029Zj7bUR@132.216.98.100
DTSTAMP:20260710T222840Z
DESCRIPTION: \nA word of caution regarding stabilized weights when using ma
 rginal structural models to estimate the effect of exposure history\nDenis
  Talbot\, PhD\n	Biostatistics Professor\, Department of Social and Preventi
 ve Medicine\, Laval University\nAbstract:\nMarginal structural models (MSM
 s) are commonly used to estimate the causal effect of a time-varying expos
 ure in presence of time-dependent confounding. These models have gained va
 st popularity over the recent years\, in part due to their ease of impleme
 ntation. Indeed\, the most common implementation of MSMs consists of fitti
 ng a weighted outcome model (e.g. a linear regression)\, where observation
 s are weighted according to their inverse-probability-of-treatment weights
 . Users must however be wary of a number of subtle issues regarding MSMs t
 hat might impact the validity of the inferences. In this talk\, I first in
 troduce MSMs and then present one such issue: the stabilization of the wei
 ghts. This stabilization is usually recommended to reduce variability\, bu
 t I provide a simple example and some simulation results where it is also 
 observed to yield biased inferences.\nBio:\nhttps://sites.google.com/site/
 denistalbotfmed1/home\n
DTSTART:20150922T193000Z
DTEND:20150922T203000Z
LOCATION:Room 24\, Purvis Hall\, CA\, QC\, Montreal\, H3A 1A2\, 1020 avenue
  des Pins Ouest
SUMMARY:Biostatistics Seminar
URL:/channels/event/biostatistics-seminar-254824
END:VEVENT
END:VCALENDAR