摘要
在定数截尾试验下,假设Pareto分布尺度参数α为已知,当形状参数θ的先验分布在分布族Γ_(1)和Γ_(2)上变化时,研究了在对称熵损失函数下,Pareto分布形状参数θ的稳健Bayes估计——条件Γ-minimax估计问题。并利用Monte-Carlo方法进行了模拟,结果表明,条件Γ-minimax估计具有较好的后验稳健性。
Based on the type-Ⅱ censoring life test,assuming that the prior distribution of parameterθvaries atΓ_(1 )andΓ_(2),conditionalΓ-minimax(CGM)estimates forθunder the symmetric entropy loss were solved whenαis known.CGM estimation is a robust Bayes estimation.Monte-Carlo simulations are performed to analyze the estimates of CGM.The results show that CGM has posterior robustness.
作者
周巧娟
ZHOU Qiaojuan(Department of Basic Courses,Nanjing Audit University Jinshen College,Nanjing 210023,China)
出处
《邵阳学院学报（自然科学版）》
2022年第1期8-13,共6页
Journal of Shaoyang University：Natural Science Edition
基金
江苏省高校哲学社会科学研究项目(2021SJA2279)。
