Parameter changes at unknown times in non-linear regression

被引:0
|
作者
Jandhyala, VK [1 ]
Al-Saleh, JA
机构
[1] Washington State Univ, Dept Pure & Appl Math, Pullman, WA 99164 USA
[2] Washington State Univ, Program Stat, Pullman, WA 99164 USA
[3] Kuwait Univ, Dept Stat, Kuwait 13060, Kuwait
关键词
Bayes-type tests; Brownian motion; change-point; non-linear regression;
D O I
暂无
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Introducing the change-point problem for the non-linear regression model, we derive Bayes-type statistics for tests of parameter changes at unknown times in the non-linear parameters. We then establish asymptotic distribution theory, mainly, for the two-sided statistics. Throughout the paper, we emphasize the specific case of an exponential type non-linear regression model. The methodology is illustrated through data on pre-school boys' weight/height ratio. Copyright (C) 1999 John Wiley & Sons, Ltd.
引用
收藏
页码:711 / 724
页数:14
相关论文
共 50 条