Activity Number:
|
511
- High-Dimensional Data Analytics: Theory and Applications
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 2, 2017 : 10:30 AM to 12:20 PM
|
Sponsor:
|
International Chinese Statistical Association
|
Abstract #324279
|
|
Title:
|
Optimal Designs for Binary Response Models with Multiple Nonnegative Variables
|
Author(s):
|
Mong-Na Lo Huang* and Shih-Hao Huang and Cheng-Wei Lin
|
Companies:
|
National Sun Yat-Sen Univ and National Sun Yat-sen University and National Sun Yat-sen University
|
Keywords:
|
Essentially complete class ;
logic model ;
Probit model ;
Schur ordering
|
Abstract:
|
In this work, we investigate optimal design problems for binary response models with multiple nonnegative explanatory variables which occur in many scientific experiments. We construct an essentially complete class of optimal designs with a simple structure for phi-p-optimality. Then we identify explicitly locally D-optimal designs within this class. Later we obtain D-optimal designs for design spaces with some more restrictions, such as two-level factorial or mixture experiments.
|
Authors who are presenting talks have a * after their name.