Some results on multi-level factorial designs with dependent observations

被引：10

作者：

Martin, RJ

Eccleston, JA

Jones, G

机构：

[1] Univ Sheffield, Sch Math & Stat, Sheffield S3 7RH, S Yorkshire, England

[2] Univ Queensland, Dept Math, Brisbane, Qld 4072, Australia

[3] Massey Univ, Dept Stat, Palmerston North, New Zealand

来源：

JOURNAL OF STATISTICAL PLANNING AND INFERENCE | 1998年 / 73卷 / 1-2期

关键词：

factorial design; optimal design; spatial dependence; temporal dependence;

D O I：

10.1016/S0378-3758(98)00054-8

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：

In many situations, observations in factorial experiments may be dependent. Run orders are then needed that result in efficient estimates of contrasts of interest. The theory is reasonably well understood for two-level designs under one-dimensional dependence. In this paper, we extend some results to multi-level factorial designs under one-dimensional dependence. (C) 1998 Elsevier Science B.V. All rights reserved.

引用

页码：91 / 111

页数：21

共 16 条

[1]

BAILEY RA, 1984, J ROY STAT SOC B MET, V46, P323

[2]

BAILEY RA, 1992, STAT SINICA, V2, P393

[3]

CHENG CS, 1991, BIOMETRIKA, V78, P325

[4]

CHENG CS, 1998, IN PRESS ANN STAT