Activity Number:
|
47
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, August 11, 2002 : 4:00 PM to 5:50 PM
|
Sponsor:
|
General Methodology
|
Abstract - #301297 |
Title:
|
Optimal Design for the Proportional Odds Model
|
Author(s):
|
Inna Perevozskaya*+ and William Rosenberger and Linda Haines
|
Affiliation(s):
|
Merck Biometrics Research Laboratories and University of Maryland, Baltimore County and University of Natal Pietermaritzburg
|
Address:
|
RY 34-A304, PO Box 2000, Rahway, New Jersey, 07065, USA
|
Keywords:
|
bayesian optimal design ; constrained optiamal design ; locally optimal design ; ethics ; ordinal response
|
Abstract:
|
We construct locally optimal designs for the proportional odds model for ordinal data. While we investigate the standard $D$-optimal design, we also investigate optimality criteria for the simultaneous estimation of multiple quantiles, namely $D_A$-optimality and the omnibus criterion. The design of experiments for the simultaneous estimation of multiple quantiles is important in both toxic and effective dose studies in medicine. As with $c$-optimality in the binary response problem, we find that there are distinct phase changes when exploring extreme quantiles that require additional design points. We also investigate relative efficiencies of the criteria.
|