摘要
对于两个给定的指数分布总体,基于截尾样本,讨论其均值θi(i=1,2)极大似然估计的比较问题。分别就定时截尾样本和定数截尾样本,给出θi(i=1,2)极大似然估计满足随机序的充分条件。最后,针对双参数指数分布,基于逐次结尾样本,给出参数极大似然估计满足随机序的充分条件,并加以证明。
For two given exponential distribution populations, the comparison problem of the maximum likelihood estimation of the mean θi(i=1,2) is discussed based on the truncated samples. The timed censored samples and the censored samples are given respectively, and the sufficient conditions for the θi(i=1,2) maximum likelihood estimation to satisfy the random order are given. Finally, for the two-parameter exponential distribution, based on the successively ending samples, the sufficient conditions for the parameter maximum likelihood estimation to satisfy the random order are given and proved.
作者
蔡明书
李树有
宓颖
CAI Ming-shu;LI Shu-you;MI Ying(College of Science,Liaoning University of Technology,Jinzhou 121001,China)
出处
《辽宁工业大学学报（自然科学版）》
2019年第5期288-294,共7页
Journal of Liaoning University of Technology(Natural Science Edition)
关键词
指数分布
极大似然估计(MLE)
Ⅰ型截尾样本
Ⅱ型截尾样本
Ⅱ型逐次截尾样本
exponential distribution
maximum likelihood estimation (MLE)
type I censored sample
type Ⅱ censored sample
type Ⅱ successive censored sample
