BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A comparison of three Bayesian approaches for constructing model r
 obust designs - Agboto\, V (Meharry Medical College)
DTSTART:20110902T140000Z
DTEND:20110902T143000Z
UID:TALK32626@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:While optimal designs are commonly used in the design of exper
 iments\, the optimality of those designs frequently depends on the form of
  an assumed model. Several useful criteria have been proposed to reduce su
 ch dependence\, and efficient designs have been then constructed based on 
 the criteria\, often algorithmically. In the model robust design paradigm\
 , a space of possible models is specified and designs are sought that are 
 efficient for all models in the space. The Bayesian criterion given by DuM
 ouchel and Jones (1994)\, posits a single model that contains both primary
  and potential terms. In this article we propose a new Bayesian model robu
 stness criterion that combines aspects of both of these approaches. We the
 n evaluate the efficacy of these three alternatives empirically. We conclu
 de that the model robust criteria generally lead to improved robustness\; 
 however\, the increased robustness can come at a significant cost in terms
  of computing requirements. \n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR