定数截尾场合Pareto分布形状参数的最优置信区间  

The Optimal Confidence Interval of Pareto Distribution Shape Parameter Based on Type-Ⅱ Censoring Test

作　　者：刘璐 李云飞[1] LIU Lu;LI Yunfei(School of Mathematics and Information, China West Normal University, Nanchong, Sichuan 637009, China)

机构地区：[1]西华师范大学数学与信息学院,四川南充637009

出　　处：《内江师范学院学报》2021年第2期45-48,共4页Journal of Neijiang Normal University

基　　金：西华师范大学英才科研基金项目(17YC381)。

摘　　要：首先采用传统方法,在置信水平为1-α=0.95时,求得定数截尾寿命试验下Pareto分布形状参数的置信区间,分析发现该置信区间并非最优置信区间.然后给出了满足定数截尾样本下参数最优置信区间的条件,进一步求解出Pareto分布形状参数的最优置信区间.最后将其与传统方法求得的置信区间进行对比分析,发现当失效个数r≤11时,具有一致最小平均长度性质的置信区间(最优置信区间)作为形状参数的置信区间将会使估计精度得到提高.First,the confidence interval of Pareto distribution shape parameter under life test of type-Ⅱ censoring case when the confidence level is 1-α =0.95 is obtained by use of the traditional method,and it is discovered that this confidence interval is not optimal.Then,the conditions that satisfy the confidence interval of the optimal parameters under type-Ⅱ censoring test are given,and the optimal confidence interval of the shape parameters of Pareto distribution is further worked out.Finally,it is then compared with the confidence interval obtained by the traditional method,and it is found that whenr#11,the use of confidence interval with the nature of consistent minimum average length(the optimal confidence interval)as the confidence interval of shape parameter will improve the estimation accuracy.

关 键 词：定数截尾试验 PARETO分布 最优置信区间 形状参数 

分 类 号：O212[理学—概率论与数理统计]