BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Equivalence of prospective and retrospective likelihood methods in
  case-control studies - Ana-Maria Staicu (Bristol)
DTSTART:20080222T140000Z
DTEND:20080222T150000Z
UID:TALK10284@talks.cam.ac.uk
CONTACT:8048
DESCRIPTION:Case-control studies are typically described by a retrospectiv
 e likelihood\, which is based on the probability of exposure given the dis
 ease status. Nevertheless\, analyzing case-control studies by prospective 
 likelihood methods\, which have the advantage of decreasing the dimension 
 of the nuisance parameter involved\, is the default approach undertaken by
  most practitioners. In this talk we address the equivalence of the prospe
 ctive and retrospective analyses in both Bayesian and frequentist settings
 . From a Bayesian perspective we give prior density conditions for the cor
 responding posterior distributions of the log-odds ratio parameter resulti
 ng from the prospective and retrospective likelihoods to be identical. The
 se conditions give Seaman and Richardson's priors when it is required that
  the nuisance parameters are independent of one another. From a frequentis
 t perspective we show the Cox-Reid modified profile likelihoods derived fr
 om the prospective and retrospective models yield the same log-odds estima
 tor up to second order. We examine the equivalence of the two models at hi
 gher order via a simulation study: for the prospective analysis we use Bar
 tlett correction\, Skovgaard's statistic and a new approach based on match
 ing priors with approximate Bayesian inference\, while for the retrospecti
 ve analysis we use Bartlett correction. \n
LOCATION:MR12\, CMS\, Wilberforce Road\, Cambridge\, CB3 0WB
END:VEVENT
END:VCALENDAR