基于成败型数据步加试验的Weibull分布的参数估计  

Parameter Estimation of Weibull Distribution with Binary Data of the Step-Stress Accelerated Life Testing

作　　者：赵竑愷 

机构地区：[1]哈尔滨理工大学理学院,黑龙江哈尔滨

出　　处：《统计学与应用》2017年第3期361-366,共6页Statistical and Application

摘　　要：在步加应力寿命试验下,对寿命服从双参数Weibull分布的产品,通过引入该产品在此步加试验下,其“寿命”的分布函数,用极大似然估计法,基于成败型数据得到了有关参数的似然方程组,并用多元二分法给出该超越方程组的求解算法,从而通过加速方程求得正常应力水平下寿命分布的参数估计。For the product with the double-parameter Weibull distribution, the distribution of the “life” of the product based on the step-stress accelerated life testing is obtained. Then I use the maximum likelihood estimation to give the likelihood equations of the parameters based on the binary data. And the algorithm of transcendental equations is given by the multivariate dichotomy, and the parameter estimation of the life distribution under the normal stress is obtained by the accelera-tion equation.

关 键 词：WEIBULL分布 步加试验 成败型数据 极大似然估计 超越方程组 二分法 

分 类 号：O21[理学—概率论与数理统计]