BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Information in adaptive optimal design with emphasis on the two st
 age case - Flournoy\, N (Missouri-Columbia)
DTSTART:20110830T103000Z
DTEND:20110830T110000Z
UID:TALK32560@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:In 1963\, Box and Hunter\, followed by many others\, recommend
 ed selecting sequential treatments to maximize the increment of some infor
 mation measure (e.g.\, the determinant of the Fisher information matrix). 
 Under nonlinear regression models\, because information is a function of u
 nknown parameters\, such increments must be estimated\; and the informatio
 n from different stages is not independent. To explore the accrual of info
 rmation in adaptive designs\, we study a basic one parameter nonlinear reg
 ression model with additive independent normal errors. The stage 1 treatme
 nt is taken to be fixed\, the treatment allocation rule for stage 2 is tak
 en to be a unique function of maximum likelihood estimates derived from st
 age 1 data. Although conditioning on the design is common in data analyses
 \, we show in this scenario\, that conditioning on the stage 2 treatment i
 s equivalent to conditioning on the stage 1 data. This raises questions ab
 out the role of conditioning in the analysis o f adaptive designs. We also
  explore the efficiency conducting studies in stages and the effect of all
 ocating different proportions of subjects to stage 1 versus stage 2. \n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR