A distribuição beta Weibull inversa generalizada: desenvolvimentos e aplicações

• Autor(a) principal: ROGES, Danielle Loureiro
• Data de Publicação: 2011
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFRPE
• Texto Completo: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/4869
• topic: Beta generalização; Estimação; Inferência; Sobrevivência; Verossimilhança; Weibull; Beta generalization; Estimation; Inference; Likelihood; Survival; CIENCIAS EXATAS E DA TERRA::PROBABILIDADE E ESTATISTICA

The Generalized InverseWeibull distribution (WIG), proposed by Gusmão et al. (2009), which has the ability to model hazard function with unimodal shape, which are quite common in biological and reliability studies, is generalized by a beta. It has, as proposed, the Beta Generalized Inverse Weibull distribution (BGIW) with five parameters and failure rate decreasing and unimodal. The distribution BWIG has WIG as a particular case. A comprehensive treatment of mathematical properties of BWIG is provided, and found the expressions for their generating functions of moments and the certain generalized r-th moment. There was also a discussion of maximum likelihood estimation and the expressions for the elements of observed information matrix. The distribution of Log-Beta Generalized Inverse Weibull and its respective regression also been developed. Also were found the Entropy of Shanonn and the Order Statistics of the distribution BWIG. A comparison of the results provided by the WIG and BWIG distributions was performed by modeling a data set agrarian.