BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Functional uniform prior distributions for nonlinear regression - 
 Bornkamp\, B (Novartis)
DTSTART:20110819T084500Z
DTEND:20110819T093000Z
UID:TALK32417@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:In this talk I will consider the topic of finding prior distri
 butions in nonlinear modelling situations\, that is\, when a major compone
 nt of the statistical model depends on a non-linear function. Making use o
 f a functional change of variables theorem\, one can derive a distribution
  that is uniform in the space of functional shapes of the underlying nonli
 near function and then back-transform to obtain a prior distribution for t
 he original model parameters. The primary application considered in this a
 rticle is non-linear regression in the context of clinical dose-finding tr
 ials. Here the so constructed priors have the advantage that they are para
 metrization invariant as opposed to uniform priors on parameter scale and 
 can be calculated prior to data collection as opposed to the Jeffreys prio
 r. I will investigate the priors for a real data example and for calculati
 on of Bayesian optimal designs\, which require the prior distribution to b
 e available before data collection has started (so that classical objectiv
 e priors such as Jeffreys priors cannot be used). \n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR