摘要
文章针对两参数Pareto分布的Bayes估计问题,在尺度参数λ已知的情况下,运用Bayes定理,得到复合MLINEX对称损失函数下形状参数θ的Bayes估计,并证明其是可容许的。最后在给定三类不同超参数的先验密度函数下,给出E-Bayes估计和多层Bayes估计,且得到E-Bayes估计具有保序性。
Aiming at the Bayes estimation problem of the two-parameter Pareto distribution, and under the condition of knowing scale parameter λ, this paper uses Bayes theorem to obtain the Bayes estimation of shape parameter θ under compound MLINEX symmetric loss function, and proves it to be admissible. Finally, the paper offers the E-Bayes estimation, multilayer Bayes estimation under the super prior density function of the three kinds of different parameters, and also finds E-Bayes estimation has order relation.
作者
朱宁
刘庆华
农以宁
蒋东云
Zhu Ning, Liu Qinghua, Nong Yining, Jiang Dongyun(School of Mathematics and Computing Science, Guilin University of Electronic Technology, Guilin Guangxi 541004, Chin)
出处
《统计与决策》
CSSCI
北大核心
2018年第5期11-15,共5页
Statistics & Decision
基金
国家科技支撑计划项目(2015BAL04B0305)
广西科技重点研发计划项目(桂科AB16380321)
桂林电子科技大学研究生教育创新计划资助项目(2016YJCX48)