Document Details

Document Type : Thesis 
Document Title :
ON PROBABILITY DISTRIBUTION OF FRAILTY MODELS
حول التوزيعات الاحتمالية لنماذج الضعف
 
Subject : Faculty of Sciences > Statistics department 
Document Language : Arabic 
Abstract : On Probability Distributions of Frailty Models Amani Abdul Aziz AL-Ahmadi Supervised By Dr. Samia Abbas Adham The survival analysis of lifetime data plays an important role in medicine, epidemiology, biology, demography, economics, engineering and other fields. The Frailty models are one of the most commonly used in survival analysis. Frailty models have become very popular during the last three decades and their applications are numerous. A frailty model is a heterogeneity model where the frailties are assumed to be individual or shared. Frailty model is an extension of the Coxs proportional hazards model, which was first introduced in Cox (1972). The main goal of this thesis is to compare the Cox’s model with its extension, when introducing gamma and inverse Gaussian frailty models. Classical and semi-parametric techniques were used for statistical inference. A real data set is applied for the models in order to deal with model comparison. The AIC (The Akaike Information Criteria) and BIC (Bayesian Information Criteria) were computed. It has been found that gamma frailty model is the best model fit this data set. Then the inverse Gaussian frailty model provides a better fit of this data set than Cox’s model. Moreover, Wald test used to test heterogeneity parameter. It has been concluded that the gamma frailty model is better than inverse Gaussian frailty model. 
Supervisor : DR. 
Thesis Type : Master Thesis 
Publishing Year : 1436 AH
2015 AD 
Added Date : Thursday, November 5, 2015 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
أماني عبد العزيز الاحمديAL-AHMADI, AMANI ABDUL AZIZResearcherMaster 

Files

File NameTypeDescription
 38126.pdf pdf 

Back To Researches Page