Maximum likelihood estimation under a finite mixture of generalized exponential distributions based on censored data

被引:0
|
作者
Saieed F. Ateya
机构
[1] Taif University,Department of Mathematics & Statistics, Faculty of Science
[2] Faculty of Science,Department of Mathematics
[3] Assiut University,undefined
来源
Statistical Papers | 2014年 / 55卷
关键词
Generalized exponential distribution; Weibull distribution; Kolmogorov–Smirnov test; Identifiability of finite mixture distributions; Generalized right-censored failure times; Random right-censored failure times; Type-I and type-II censoring; EM algorithm; 62F10; 62N01; 62N02;
D O I
暂无
中图分类号
学科分类号
摘要
In this paper, the identifiability of a finite mixture of generalized exponential distributions (GE(τ, α)) is proved and the maximum likelihood estimates (MLE’s) of the parameters are obtained using EM algorithm based on a general form of right-censored failure times. The results are specialized to type-I and type-II censored samples. A real data set is introduced and analyzed using a mixture of two GE(τ, α) distributions and also using a mixture of two Weibull(α, β) distributions. A comparison is carried out between the mentioned mixtures based on the corresponding Kolmogorov–Smirnov (K–S) test statistic to emphasize that the GE(τ, α) mixture model fits the data better than the other mixture model.
引用
收藏
页码:311 / 325
页数:14
相关论文
共 50 条