Likelihood, credible and confidence intervals for parameters in complex models

被引:1
|
作者
Aitkin, Murray [1 ]
机构
[1] Univ Melbourne, Sch Math & Stat, Melbourne, Vic, Australia
来源
关键词
Likelihood; Pivot; Confidence distribution; Asymptotic coverage; Non-informative prior;
D O I
10.1007/s40300-018-0144-4
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
Recent papers have discussed general procedures with complex models to obtain confidence interval statements for a parameter of interest in the presence of nuisance parameters. This paper discusses the role of the likelihood in these procedures, and points out the simplicity of the Bayesian credible interval approach to the same models.
引用
收藏
页码:305 / 320
页数:16
相关论文
共 50 条