|
Activity Number:
|
21
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #303674 |
|
Title:
|
Design and Analysis of Computer Experiments with Grouped Factors
|
|
Author(s):
|
Jin Xu*+ and Peter Z.G. Qian
|
|
Companies:
|
East China Normal University and University of Wisconsin-Madison
|
|
Address:
|
, Shanghai, 200241, China
|
|
Keywords:
|
Computer experiments ; grouped space-filling design ; grouped factors
|
|
Abstract:
|
Computer experiments are widely used in engineering and sciences for studying real world systems. Grouping among factors naturally arise in many computer computers. A statistical framework for the design and analysis of such experiments is developed. We propose a new type of design called grouped space-filling design in which the points have good space-filling properties both at the individual factor level and at the group level. Some new Gaussian process models that can accommodate grouped factors are also developed. Several examples are provided to illustrate the proposed methods.
|
