A New Distribution for Extreme Values: Regression Model, Characterizations and Applications

Journal of Data Science

article

H.M. Yousof¹  S.M.A. Jahanshahi²  T.G. Ramires³  G.R. Aryal⁵  G.G. Hamedani⁶ 

[1] Department of Statistics, Mathematics and Insurance, Benha University;Faculty of Mathematics, University of Sistan and Baluchestan;Department of Exact Sciences, University of Sao Paulo;Institute for Biostatistics and Bioinformatics, University of Haslet;Department of Mathematics, Statistics and CS, Purdue University Northwest;Department of Mathematics, Statistics and CS, Marquette University

关键词: Extreme value distribution;    Order statistics;    Parameter estimation;   

DOI  :  10.6339/JDS.201810_16(4).00002

学科分类：土木及结构工程学

来源: JDS

【 摘 要 】

A new four parameter extreme value distribution is defined and studied. Various structural properties of the proposed distribution including ordinary and incomplete moments, generating functions, residual and reversed residual life functions, order statistics are investigated. Some useful characterizations based on two truncated moments as well as based on the reverse hazard function and on certain functions of the random variable are presented. The maximum likelihood method is used to estimate the model parameters. Further, we propose a new extended regression model based on the logarithm of the new distribution. The new distribution is applied to model three real data sets to prove empirically its flexibility.

【 授权许可】

CC BY