Comparison of designs for computer experiments

被引:69
作者
Bursztyn, D
Steinberg, DM [1 ]
机构
[1] Tel Aviv Univ, Dept Stat & Operat Res, Raymond & Beverly Sackler Fac Exact Sci, IL-69978 Tel Aviv, Israel
[2] Ashkelon Coll, Ashqelon, Israel
基金
以色列科学基金会;
关键词
alias matrix; entropy criterion; Latin hypercube designs; Lattice designs; maximin designs; mean squared error criterion; random field regression; rotation designs;
D O I
10.1016/j.jspi.2004.08.007
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
For many complex processes laboratory experimentation is too expensive or too time-consuming to be carried out. A practical alternative is to simulate these phenomena by a computer code. This article considers the choice of an experimental design for computer experiments. We illustrate some drawbacks to criteria that have been proposed and suggest an alternative, based on the Bayesian interpretation of the alias matrix in Draper and Guttman (Ann. Inst. Statist. Math. 44 (1992) 659). Then we compare different design criteria by studying how they rate a variety of candidate designs for computer experiments such as Latin hypercube plans, U-designs, lattice designs and rotation designs. (c) 2004 Elsevier B.V. All rights reserved.
引用
收藏
页码:1103 / 1119
页数:17
相关论文
共 32 条
[1]  
[Anonymous], [No title captured]
[2]  
Bates RA, 1996, J ROY STAT SOC B MET, V58, P77
[3]  
BEATTIE SD, 1997, P SPES AM STAT ASS
[4]  
BOX GEP, 1959, J AM STAT ASSOC, V54, P622
[5]   CHOICE OF A SECOND ORDER ROTATABLE DESIGN [J].
BOX, GEP ;
DRAPER, NR .
BIOMETRIKA, 1963, 50 (3-4) :335-&
[6]   Rotation designs: orthogonal first-order designs with higher order projectivity [J].
Bursztyn, D ;
Steinberg, DM .
APPLIED STOCHASTIC MODELS IN BUSINESS AND INDUSTRY, 2002, 18 (03) :197-206
[7]   Rotation designs for experiments in high-bias situations [J].
Bursztyn, D ;
Steinberg, DM .
JOURNAL OF STATISTICAL PLANNING AND INFERENCE, 2001, 97 (02) :399-414
[8]   BAYESIAN PREDICTION OF DETERMINISTIC FUNCTIONS, WITH APPLICATIONS TO THE DESIGN AND ANALYSIS OF COMPUTER EXPERIMENTS [J].
CURRIN, C ;
MITCHELL, T ;
MORRIS, M ;
YLVISAKER, D .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1991, 86 (416) :953-963
[9]   TREATING BIAS AS VARIANCE FOR EXPERIMENTAL-DESIGN PURPOSES [J].
DRAPER, NR ;
GUTTMAN, I .
ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS, 1992, 44 (04) :659-671
[10]   SOME APPLICATION OF NUMBER-THEORETIC METHODS IN STATISTICS [J].
FANG, KT ;
WANG, Y ;
BENTLER, PM .
STATISTICAL SCIENCE, 1994, 9 (03) :416-428