Document Details

Document Type : Thesis 
Document Title :
AN EXTENSIVE STUDY ON SOME LIFETIME DISTRIBUTIONS WITH THEIR APPLICATIONS
دراسة مكثفة حول بعض توزيعات الحياة و تطبيقاتها
 
Subject : Faculty of Science 
Document Language : Arabic 
Abstract : The parameter estimation is a challenging task especially in the case of mixture distributions, because we can not find the parameters in closed form. Here, we pay more attention to some mixture lifetime distributions that have been proposed recently. We aim to consider two models and use different methods of estimation under censoring data. In this study, the problem of classical estimation based on unknown parameters of exponential-geometric(EG) distribution under progressive type-II censored data is considered. It is usually, we find the maximum likelihood estimators(MLEs), bias and mean squared error(MSE) of parameters of EG distribution. However, in our situation, it is not an easy task to get the estimators of the involved parameters in closed forms, therefore, we propose to use the Expectation Maximization(EM) algorithm. The asymptotic variance of the MLEs of the EG parameters is computed along with the asymptotic confidence intervals(CI). A simulation study is provided to evaluate the performances of EG model under progressive type-II censored data. Furthermore, the problem of classical and Bayesian estimations based on unknown parameters of the complementary exponential-geometric(CEG) distribution under progressive type-II censored data is considered. The maximum likelihood(ML) and Bayes estimates of parameters of CEG distribution are also of interest to us. Also, we derive Bayes estimates based on squared error loss function(SELF). In order to compute Bayes estimate, Markov Chain Monte Carlo(MCMC) techniques are used. The CI intervals of the MLEs, Highest Posterior Density(HPD) intervals, bias and MSE are obtained. Comparisons are made between ML and Bayesian methods through a simulated and real data sets. 
Supervisor : Dr. Aisha Fayomi 
Thesis Type : Master Thesis 
Publishing Year : 1440 AH
2019 AD 
Added Date : Monday, June 17, 2019 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
حمدة طارش الشمريAl-Shamari, Hamdah TareshResearcherMaster 

Files

File NameTypeDescription
 44481.pdf pdf 

Back To Researches Page