指数分布串—并联混合系统产品的统计分析  

The Statistical Analysis in a Mixed System of Series and Parallel Connection for Exponential Distribution

作　　者：袁芳[1] 王蓉华[2] 徐晓岭[3] 

机构地区：[1]辉瑞(中国)研究开发有限公司,上海201203 [2]上海师范大学数理信息学院,上海200234 [3]上海对外贸易学院商务信息学院,上海201600

出　　处：《大学数学》2011年第4期29-35,共7页College Mathematics

基　　金：上海市教委基金(06MS009;CL200517);上海市重点学科(T0401);国家自然科学基金(10571057);上海市科委科技项目(075105118);科学计算上海高校重点实验室;2007年度市教委重点课程(5Z1206);2007年度统计学专业建设(5Z1501)

摘　　要：给出单元寿命服从同一指数分布的串-并联混合系统产品参数的矩估计和极大似然估计,并通过大量Monte-Carlo模拟比较了估计的精度,得到在样本容量小于35时矩估计优于极大似然估计,而样本容量不小于35时极大似然估计优于矩估计.另外,还给出了参数的精确区间估计与近似区间估计,并通过大量Monte-Carlo模拟考察了区间估计的精度.The paper presents the moment estimate and the maximum likelihood estimate of the parameters of a product which is a mixed system of series and parallel connection in which the lifetime of every single unit submitted to the same exponential distribution.Moreover,the precisions of the above point estimates are compared by performing a great deal of Monte-Carlo simulation.It can be concluded that the moment estimate is better than the maximum likelihood estimate when the sample size is less than 35 while it＇s the opposite when the sample size is not less than 35.Furthermore,the precise interval estimate and the approximate interval estimate of the parameters are presented and a large amount of Monte-Carlo simulation is conducted to investigate the precisions of the above interval estimates.

关 键 词：串-并联混合系统 指数分布 矩估计 极大似然估计 精确区间估计 近似区间估计 

分 类 号：O213.2[理学—概率论与数理统计]