Document Details
|
Document Type |
: |
Thesis |
|
Document Title |
: |
On Mixtures of Exponentiated Exponentil Destributioons. حول خلائط للتوزيعات الاسي أسية. |
|
Subject |
: |
Girls' College of Education in Jeddah |
|
Document Language |
: |
Arabic |
|
Abstract |
: |
The exponentiated exponential (EE) or generalized exponential (GE) distribution is introduced as a special case of the exponentiated Weibull with three parameters.
The significance of this distribution can be used effectively to analyze the data life times, as well as his perfect interpretations of the physical, that it is a parallel system of n components - in the sense that the system works if at least one of the components works - and whether the distributions of lifetime of these components are exponential random variables with independent and identical distributions, then the distribution of lifetime of this system follows the exponentiated exponential distribution.
This thesis concentrated on the mixture of two exponentiated exponential distributions. The main aims of this study are to present the mixture under study, discuss it's properties, study its relations with other distributions, estimate the two shape parameters by different methods of non-Bayesian estimation, and these methods are: maximum likelihood method for complete samples and censored samples type I, moments method, percentiles method, least squares and weighted least squares, then compare the estimators and determined which kind more better than the others using Monte Carlo simulation study, estimate the two parameters, survival and failure rate functions by Bayesian method, and compare the maximum likelihood estimators and Bayesian estimators based on two kind of loss functions, that is, quadratic loss function and linear-exponential (LINEX) loss function using Lindley's approximation by Monte Carlo simulation study. Study the order statistics and record values of the mixture were also aims in this thesis, the study including the inferences based on order statistics and record values, fisher information matrices with n order statistics and record values well be drive.
Applying the goodness of fit tests based of empirical distribution function and get the critical values tables of the modified Kolmogorov-Smirnov (KS), Anderson-Darling (AD), Cramer–von Mises (CvM) goodness-of-fit tests for mixture using different sizes of complete samples and Type II censored samples with censoring fractions 80%, 90% , in case two shape parameters unknown ,
Mathematica 4.0 computer program was used to get tables, figures and numerical results. |
|
Supervisor |
: |
dr.Shawki Ahmed |
|
Thesis Type |
: |
Doctorate Thesis |
|
Publishing Year |
: |
1431 AH
2010 AD |
|
Added Date |
: |
Saturday, March 13, 2010 |
|
Researchers
| مها علي الدحلان | AL- Dahlan, Maha Ali | Researcher | Doctorate | |
|
Back To Researches Page
|