|
Activity Number:
|
24
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #306572 |
|
Title:
|
Locally D-Optimal Designs for Pharmacokinetics Compartmental Models
|
|
Author(s):
|
Xin Fang*+
|
|
Companies:
|
University of Illinois at Chicago
|
|
Address:
|
702 S. Lytle Street, CHICAGO, IL, 60607,
|
|
Keywords:
|
d-optimal ; pharmacokinetics ; compartmental models
|
|
Abstract:
|
Locally D-optimal designs are investigated for an open one-compartment model with zero order input and first order elimination and a general compartmental model with 2n parameters. These nonlinear models have been used and studied in pharmaceutical industries and other research organizations. We allow an additive error which follows a Normal distribution. The LD design for the two-parameter one compartmental model are found to be saturated with the explicit two design points. An upper bound of the number of points in the support of the LD design for the general compartmental model with 2n parameters is updated. A sufficient condition for the inclusion of the sampling time 0 to be in the LD design support of the general compartmental model is provided. Robust designs are investigated. Simulation results suggest that LD designs are relatively robust designs in regard to the nominal values.
|
