Document Details

Document Type : Thesis 
Document Title :
Bivariate Exponentiated Pareto Distributions Based on Mixture and Copula
توزيعات باريتو الأسي ثنائية المتغيرات على أساس الرابطة والخليط
 
Subject : Faculty of Sciences 
Document Language : Arabic 
Abstract : The exponentiated Pareto distribution has been used quite effectively to model many lifetime data. Constructing and studying bivariate probability distributions are of great interests of many statisticians. A popular and flexible way to derive different bivariate lifetime distributions using copula functions. In this Thesis, two new bivariate exponentiated Pareto distributions are introduced. The first proposed bivariate distribution is constructed based on Gaussian copula with exponentiated Pareto distribution as marginals and the second bivariate distribution is constructed based on M mixture representation and Gaussian copula. Several properties of the proposed bivariate distributions can be obtained using the Gaussian copula property. Different methods of estimation of the unknown parameters of proposed bivariate distributions are considered. The Markov Chain Monte Carlo technique has been used to compute the Bayesian estimates based on squared error loss function. Moreover, Monte Carlo simulation study is used to investigate and compare the different estimates for different sample sizes and for different values of the Gaussian copula parameter. Simulation results showed that Bayesian method in most cases provides more accurate estimates compared to other methods. In addition, the results based on mean square error showed that second bivariate distribution provides more accurate estimates compared to the first bivariate distribution. Finally, real data set is analyzed and the results showed that the proposed distributions gave more satisfactory performance compared to some other very well-known distributions. 
Supervisor : Dr. Lamya Baharith 
Thesis Type : Master Thesis 
Publishing Year : 1438 AH
2016 AD 
Co-Supervisor : Dr. Mervat Khalifa 
Added Date : Wednesday, January 11, 2017 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
أشواق سليم العرويALerwi, Ashwag SaleemResearcherMaster 

Files

File NameTypeDescription
 39571.pdf pdf 

Back To Researches Page