Some new results for the transmuted generalized gamma distribution

Author(s)

Saboor, Abdus ^[1] ; Khan, Muhammad Nauman ^[1] ; Cordeiro, Gauss M. ^[2] ; Pascoa, Marcelino A. R. ^[3] ; Ramos, Pedro L. ^[4] ; Kamal, Mustafa ^[1]

Affiliation

^[1] Kohat Univ Sci & Technol, Dept Math, Kohat 26000 - Pakistan

^[2] Univ Fed Pernambuco, Dept Estat, BR-50740540 Recife, PE - Brazil

^[3] Univ Fed Mato Grosso, Dept Estat, BR-78075850 Cuiaba, MT - Brazil

^[4] Univ Sao Paulo, Inst Math Sci & Comp, Sao Carlos, SP - Brazil

• Document type: Journal article
• Source: Journal of Computational and Applied Mathematics; v. 352, p. 165-180, MAY 15 2019.
• Web of Science Citations: 0

Abstract

The last decade is full on new classes of distributions that become precious for applied statisticians. Extending known distributions by adding parameters enables us to obtain more flexible models. We provide some new mathematical properties of the transmuted generalized gamma distribution defined from the family pioneered by Aryal and Tsokos (2011). We derive explicit expressions for some of its mathematical quantities. We present maximum likelihood and Bayesian estimators for the model parameters. The different estimators are compared using extensive numerical simulations. For the sake of illustration, we apply our proposed methodology to two real data sets, thus proving empirically that the current distribution is a simple alternative for lifetime data. (C) 2018 Elsevier B.V. All rights reserved. (AU)

• FAPESP's process: 17/25971-0 - Statistical inference of complex systems
• Grantee: Pedro Luiz Ramos
• Support Opportunities: Scholarships in Brazil - Post-Doctoral