Classical and Bayesian Inference of a Mixture of Bivariate Exponentiated Exponential Model

UNCW Author/Contributor (non-UNCW co-authors, if there are any, appear on document)
Indranil Ghosh (Creator)
Institution
The University of North Carolina Wilmington (UNCW )
Web Site: http://library.uncw.edu/

Abstract: Exponentiated exponential (EE) model has been used effectively in reliability, engineering, biomedical, social sciences, and other applications. In this study, we introduce a new bivariate mixture EE model with two parameters assuming two cases, independent and dependent random variables. We develop a bivariate mixture starting from two EE models assuming two cases, two independent and two dependent EE models. We study some useful statistical properties of this distribution, such as marginals and conditional distributions and product moments and conditional moments. In addition, we study a dependent case, a new mixture of the bivariate model based on EE distribution marginal with two parameters and with a bivariate Gaussian copula. Different methods of estimation for the model parameters are used both under the classical and under the Bayesian paradigm. Some simulation studies are presented to verify the performance of the estimation methods of the proposed model. To illustrate the flexibility of the proposed model, a real dataset is reanalyzed.

Additional Information

Publication
https://doi.org/10.1155/2021/5200979
Language: English
Date: 2021
Keywords
Maximum likelihood estimation, survival analysis, cure fraction, Bayesian inference, Discrete distributions, Censored data

Email this document to