BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Selection bias\, missing data and causal inference - Kate Tilling\
 , University of Bristol
DTSTART:20200228T140000Z
DTEND:20200228T150000Z
UID:TALK135973@talks.cam.ac.uk
CONTACT:Dr Sergio Bacallado
DESCRIPTION:Causal inference can be attempted using different statistical 
 methods\, each of which require some (untestable) assumptions. Common meth
 ods include multivariable regression (no unmeasured confounding)\, variati
 ons on regression such as propensity scores\, g-methods (no unmeasured con
 founding) and instrumental variables (no association between instrument an
 d outcome\, other than via the exposure). Less attention has been given to
  the impact of selection (e.g. selection into a study\, analysis of cases 
 only) or missing data (e.g. dropout from a study\, death due to other caus
 es) on causal inference. Using directed acyclic graphs (DAGs) I will show 
 some of the ways in which bias can occur due to selection or missing data.
  I will also present a recently developed method to overcome selection bia
 s using genetic data (under some assumptions). Applied work shows evidence
 \nof non-random selection into and dropout from studies including ALSPAC a
 nd UK Biobank\, and I will discuss how this might impact causal analyses u
 sing these datasets.
LOCATION:MR12
END:VEVENT
END:VCALENDAR