UMA NOVA CLASSE DE DISTRIBUICÕES WEIBULL GENERALIZADA

Main Article Content

Patrícia Silva Nascimento BARROS
Cícero Carlos Ramos de BRITO
Leandro Chaves RÊGO
Wilson Rosa OLIVEIRA
Frank Sinatra GOMES-SILVA

Abstract

The Weibull distribution is a very popular distribution for modelling lifetime data. Many modications of the Weibull distribution have been proposed in recent years. Inspired by the T-X method (ALZAATREH; LEE; FAMOYE, 2013), we propose a class of generalized Weibull distributions which generalizes the Weibull-X class. We investigate some of its mathematical properties including an expansion for the density function and explicit expressions for the ordinary and central moments, generating function and characteristic function. We discuss a particular case of the new generator and present briefly some of its structural properties. As an alternative to maximum likelihood, we performed Bootstrap simulations and corrected Bootstrap in order to compare these methods of estimation. The Bootstrap method is employed to estimate the model parameters. We t one special model of the new class to four real data sets to demonstrate the flexibility of the proposed class.

Article Details

How to Cite
BARROS, P. S. N., BRITO, C. C. R. de, RÊGO, L. C., OLIVEIRA, W. R., & GOMES-SILVA, F. S. (2017). UMA NOVA CLASSE DE DISTRIBUICÕES WEIBULL GENERALIZADA. Brazilian Journal of Biometrics, 35(2), 319–347. Retrieved from https://biometria.ufla.br/index.php/BBJ/article/view/104
Section
Articles
Author Biographies

Patrícia Silva Nascimento BARROS, Universidade Federal Rural de Pernambuco

Departamento de Ciências Exatas

Leandro Chaves RÊGO, Universidade Federal de Pernambuco

Depatamento de Estatística e Informática

Wilson Rosa OLIVEIRA, Universidade Federal Rural de Pernambuco

Departamento de Estatística e Informática

Frank Sinatra GOMES-SILVA, Universidade Federal Rural de Pernambuco

Departamento de Estatística e Informática

Most read articles by the same author(s)