双参数指数分布均值、分位数、生存概率的置信区间估计  

Confidence Interval Estimations of the Mean, p-Quantile, and Survival Probability for Two-Parameter Exponential Distribution

作　　者：王春玲 牟唯嫣 赵昕[2] WANG Chunling;MU Weiyan;ZHAO Xin(School of Science,Beijing University of Civil Engineering and Architecture,Beijing 100044;Canvard College,Beijing Technology and Business University,Beijing 101118)

机构地区：[1]北京建筑大学理学院,北京100044 [2]北京工商大学嘉华学院,北京101118

出　　处：《北京建筑大学学报》2019年第4期59-64,共6页Journal of Beijing University of Civil Engineering and Architecture

基　　金：国家自然科学基金项目(11601027,11671386)

摘　　要：给出了双参数指数分布的均值、p分位数的三种置信限,并通过Monte Carlo方法对三种方法进行模拟.同时给出了生存概率的广义枢轴量,并对其进行了模拟研究.模拟结果表明:由广义推断方法给出的置信限在中小样本的情况下表现良好,尤其在样本量较小的情况下广义置信限比基于正态修正得到的近似置信限有更好的表现,而且相比于精确方法,广义推断方法大大降低计算的复杂度;且基于广义推断方法得到的生存概率的置信区间在中小样本时接近名义置信水平,平均置信长度较小.In this paper,three confidence limits for the mean and the p-quantile of the two-parameter exponential distribution are given,and the three methods are simulated by the Monte Carlo method.At the same time,the generalized pivot amount of survival probability is given and simulated.The simulation results show that the confidence limits given by the generalized inference method performs well in the case of small and medium samples,especially in the case of small sample size,the generalized confidence ratio is better than the approximate confidence limit proposed by the normal state,and the generalized inference method greatly reduces the computational complexity compared to the exact method;Similarly,the confidence interval of the survival probability based on the generalized inference method is close to the nominal confidence level in the small and medium samples,and the average confidence length is smaller.

关 键 词：双参数指数分布 广义枢轴量 广义置信区间 MNA近似 

分 类 号：O212.4[理学—概率论与数理统计]