DS-optimal designs for random coefficient first-degree regression model with heteroscedastic errors

被引:2
|
作者
Wilk, M. [1 ,2 ]
Zaigraev, A. [1 ]
机构
[1] Nicolaus Copernicus Univ, Fac Math & Comp Sci, Ul Chopina 12-18, PL-87100 Torun, Poland
[2] Polish Acad Sci, Inst Psychol, Ul Jaracza 1, PL-00378 Warsaw, Poland
关键词
DS-optimal design; Heteroscedastic errors; FCR model; RCR model; LINEAR-MODELS; CRITERION;
D O I
10.1016/j.spl.2017.04.016
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
Optimal design problems for random coefficient regression model with heteroscedastic errors are considered. Under continuous design setting, conditions for existence of 2-point design that is as good as a given k-point design and for existence of DS-optimal design are established. (C) 2017 Elsevier B.V. All rights reserved.
引用
收藏
页码:28 / 34
页数:7
相关论文
共 32 条