BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Virtual Seminar: 'Using auxiliary variables in mediation analysis 
 to address unmeasured confounding' - Professor Richard Emsley\, NIHR Resea
 rch Professor\, Institute of Psychiatry\, Psychology and Neuroscience\, Ki
 ng’s College London
DTSTART:20201110T140000Z
DTEND:20201110T150000Z
UID:TALK153709@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:Causal mediation analysis aims to estimate natural direct and 
 natural indirect effects under clearly specified assumptions. Traditional 
 mediation analysis based on ordinary least squares (OLS) estimation assume
 s an absence of unmeasured causes to the putative mediator and outcome\, e
 ven in randomised studies. When these assumptions cannot be justified\, in
 strumental variables (IV) estimators can be used to produce an asymptotica
 lly unbiased estimator of the mediator-outcome link. However\, this bias r
 emoval comes at the cost of variance inflation and requires additional par
 ametric assumptions about the validity of the instruments. I will describe
  the statistical approaches for identification and estimation of parametri
 c regression and instrumental variable analyses\, including examples of mu
 ltiple instruments. I will introduce a new estimator that strikes a natura
 l trade-off between the unbiasedness of the IV procedure and the relativel
 y small variance of the OLS estimator. 
LOCATION:Virtual Seminar 
END:VEVENT
END:VCALENDAR