Optimal design for linear regression models in the presence of heteroscedasticity caused by random coefficients

被引：9

作者：

Grasshoff, Ulrike ^{[1
]}

Doebler, Anna ^{[2
]}

Holling, Heinz ^{[2
]}

Schwabe, Rainer ^{[1
]}

机构：

[1] Otto von Guericke Univ, Inst Math Stochast, D-39016 Magdeburg, Germany

[2] Univ Munster, Psychol Inst 4, D-48149 Munster, Germany

来源：

JOURNAL OF STATISTICAL PLANNING AND INFERENCE | 2012年 / 142卷 / 05期

关键词：

Optimal design; Linear mixed model; Random coefficient regression;

D O I：

10.1016/j.jspi.2011.11.025

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：

Random coefficients may result in heteroscedasticity of observations. For particular situations, where only one observation is available per individual, we derive optimal designs based on the geometry of the design locus. (C) 2011 Elsevier B.V. All rights reserved.

引用

页码：1108 / 1113

页数：6

共 6 条

[1] Fedorov V., 1972, Theory of Optimal Experiment
[2] Creativity in the classroom: A multilevel analysis investigating the impact of creativity and reasoning ability on GPA
Freund, Philipp Alexander
Holling, Heinz
[J]. CREATIVITY RESEARCH JOURNAL, 2008, 20 (03) : 309 - 318
[3] Mielke T, 2009, P 6 ST PET WORKSH SI, P1066
[4] Efficient sampling windows for parameter estimation in mixed effects models
Patan, Maciej
Bogacka, Barbara
[J]. MODA 8 - ADVANCES IN MODEL-ORIENTED DESIGN AND ANALYSIS, 2007, : 147 - +
[5] Sibson B., 1972, J R STAT SOC B, V34, P174
[6] Sibson R., 1972, J R STAT SOC B, V34, P181

← 1 →