期刊名称：Sankhya. Series B, applied and interdisciplinary statistics
印刷版ISSN：0976-8386
电子版ISSN：0976-8394
出版年度：2017
卷号：79
期号：2
页码：316-335
DOI：10.1007/s13571-017-0142-1
语种：English
出版社：Indian Statistical Institute
摘要：AbstractIn this paper, we propose a new compounding distribution, named the complementary Lindley-geometric distribution. It arises on a latent complementary risks scenarios where only the maximum lifetime value among all risks instead of a particular risk is observed. Its characterization and statistical properties are investigated. The maximum likelihood inference using EM algorithm is developed. Asymptotic properties of the MLEs are discussed and simulation studies are performed to assess the performance of parameter estimation. We illustrate the proposed model with a real application and it shows that the new distribution is appropriate and potential for lifetime analyses.
关键词：KeywordsEnLindley distributionGeometric distributionHazard functionMaximum likelihood estimationEM algorithmFisher information matrix
