الفهرس 
On assessing the number of components in survival finite mixture models
summaryOnly 14 pages are availabe for public view
 |  | from | 104 |  |  |
| | | from | 104 | | |
Abstract
Survival analysis is a collection of statistical procedures for data analysis for which the outcome variable of interest is time until an event occurs. Historically, survival analysis has been represented with the classical statistical distributions such as Gamma, exponential and Weibull distributions. In the heterogeneous structure of the data, the use of survival mixtures of distributions has become widespread. Survival Mixture Models SMM have been widely used as an efficient statistical tool for modeling and describing population heterogeneity in reliability and life testing applications. Both classical and Bayesian approaches were widely used in estimating the parameters of survival mixture models. However, the question of how many components should be included in a survival mixture model to adequately represent the data is still a difficult question needing a satisfactory solution. This thesis considers assessing the number of components in Survival Mixture Models SMM. Specifically, a two-component Weibull mixture of distributions will be adopted under random censoring. Weibull Mixture Models WMM are commonly used due to their mathematical flexibility and their various applications in survival and reliability analyses