BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:D-optimal designs for multinomial experiments - Russell\, K (Charl
 es Sturt University)
DTSTART:20110901T083000Z
DTEND:20110901T090000Z
UID:TALK32609@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Consider a multinomial experiment where the value of a respons
 e variable falls in\none of k classes. The k classes are not assumed to ha
 ve a hierarchical structure.\nLet ij represent the probability that the i
 th experimental unit gives a response\nthat falls in the jth class. By mod
 elling ln(ij=i1) as a linear function of the\nvalues of m predictor vari
 ables\, we may analyse the results of the experiment\nusing a Generalized 
 Linear Model.\n\nWe describe the construction of D-optimal experimental de
 signs for use in such\nan experiment. Difficulties in obtaining these desi
 gns will be described\, together\nwith attempts to overcome these obstacle
 s.\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR