摘要
混合指数威布尔分布是寿命数据分析中一个重要的统计模型.但是利用传统的矩法估计,极大似然估计等估计模型的参数比较困难.应用ECM算法,研究了混合指数威布尔分布在定数截尾数据场合下的参数估计问题,并以数值模拟验证用ECM算法来估计混合指数威布尔分布在定数截尾数据场合下的有效性.
Mixed exponential Weibull distributions is an important statistical model in life data analysis. However we may encounter enormous difficulties when we try to estimate the model by traditional methods, such as maximal likelihood and moments method estimations. In this paper,we apply ECM algorithm to study the parameter estimation of the mixed exponential Weibull model in the fully data situation,and explain it is an easy and effective method to estimate mixed exponential Weibull distributions under fixed number truncation.
出处
《数学的实践与认识》
CSCD
北大核心
2013年第22期189-195,共7页
Mathematics in Practice and Theory
基金
中青年科研基金(2012-QGY-3)
教育部春晖计划(Z2012091)
关键词
混合指数威布尔分布
定数结尾
参数估计
ECM算法
EM算法
指数威布尔分布
mixed exponentiated Weibull Distributions
fixed number truncation
parameters of estimate
ECM algorithm
EM algorithm
exponentiated Weibull Distributions
