混合删失样本下混合指数分布的估计（英文）  

Estimating Mixed Exponential Distributions Based on Hybrid Censored Samples

作　　者：田玉柱[1] 韩学峰[2] 田茂再[3] 

机构地区：[1]河南科技大学数学与统计学院,洛阳471023 [2]河南理工大学数学与信息科学学院,焦作454000 [3]中国人民大学应用统计中心,北京100872

出　　处：《应用概率统计》2017年第2期191-202,共12页Chinese Journal of Applied Probability and Statistics

基　　金：partly supported by the National Natural Science Foundation of China(Grant Nos.11501167;11271368);the Key Scientific Research Project of Henan Province Universities(Grant No.15A110025);the Doctoral Scientific Research Foundation of Henan University of Science and Technology(Grant No.4010-13480049)

摘　　要：混合删失方案是Ⅰ型和Ⅱ型删失方案的一个组合,该删失方案在寿命数据分析已非常流行.混合指数分析模型是可靠性统计中非常受欢迎的一类模型.然而,带混合删失样本的混合指数分布的参数估计问题还没有出现相关的讨论,本文将来处理这个问题.文章利用EM算法获得了参数极大似然估计的显示表达式.最后,我们开展了一些数值模拟和实际数据分析来展示所提方法的估计效果.The hybrid censoring scheme is a mixture of type-I and type-II censoring schemes. It is a popular censoring scheme in the literature of life data analysis. Mixed exponential distribution （MED） models is a class of favorable models in reliability statistics. Nevertheless, there is no much discussion to focus on parameters estimation for MED models with hybrid censored samples. We will address this problem in this paper. The EM （Expectation-Maximization） algorithm is employed to derive the closed form of the maximum likelihood estimators （MLEs）. Finally, Monte Carlo simulations and a real-world data analysis are conducted to illustrate the proposed method.

关 键 词：混合指数分布 混合删失 极大似然估计 EM算法 

分 类 号：O212.2[理学—概率论与数理统计] O213.2[理学—数学]