定时截尾样本下广义逆指数分布参数的Bayes估计  

Bayes Estimation of the Parameter for Generalized Inverted Exponential Distribution Under Type-I Censored Samples

作　　者：刘华[1] 习长新[1] Liu Hua;Xi Changxin(School of Mathematics and Physics,Jingchu University of Technology,Jingmen Hubei 448000,China)

机构地区：[1]荆楚理工学院数理学院,湖北荆门448000

出　　处：《统计与决策》2023年第15期41-47,共7页Statistics & Decision

基　　金：荆楚理工学院校级科研项目(YB202213)。

摘　　要：文章在定时截尾样本下,讨论了广义逆指数分布形状参数、可靠度和危险率的极大似然估计。基于指数先验分布,在熵损失、平方损失和Linex损失函数下分别得到形状参数、可靠度和危险率的Bayes估计,并给出了确定超参数的方法。利用数值模拟计算了估计量的各种估计均值和均方误差,研究结果表明,形状参数在熵损失和Linex损失函数下的估计精度较高;可靠度的Bayes估计整体优于极大似然估计;危险率的Bayes估计在Linex损失函数下的效果较好。This paper discusses the maximum likelihood estimation of shape parameters,reliability and hazard rate of generalized inverse exponential distribution under Type-I censored samples.Based on exponential prior distribution,Bayes estimates of shape parameters,reliability and hazard rate are obtained respectively under entropy loss,square loss and Linex loss functions,and the method of determining the super parameters is presented.Finally,the estimated mean values and mean square errors of the estimators are calculated by use of numerical simulation.The results show that the shape parameters are estimated with high accuracy under entropy loss and Linex loss function,that Bayes estimation of reliability is better than maximum likelihood estimation on the whole,and that Bayes estimation of hazard rate is better under Linex loss function.

关 键 词：定时截尾 广义逆指数分布 最大似然法 BAYES估计 数值模拟 

分 类 号：O212.8[理学—概率论与数理统计]