Open Access
January 2017 Confidence Bounds for Extreme Wind Speed Estimates: A Comparison For the Gumbel - Burr XII Distribution and Classical Extreme Value Distributions
Patrick OSATOHANMWEN, Francis O OYEGUE, Sunday Martins OGBONMWAN
Afr. J. Appl. Stat. 4(1): 259-271 (January 2017). DOI: 10.16929/ajas/2017.259.214

Abstract

Recently, the Gumbel-Burr XII (GUBXII) distribution has been introduced in Osatohanmwen et al. (2017), as a new member from the T-X family of distributions. Several properties of the new distribution were studied by the authors and an application to a real lifetime data set was carried out. In this paper, analysis of extreme wind speed frequency is undertaken. A comparison for the GUBXII distribution, generalized extreme value distribution and the Gumbel distribution in fitting weekly highest wind speed observations collected in Benin City, Nigeria for 200 weeks is performed. Results obtained indicate that the GUBXII distribution outperformed the Gumbel and the generalized extreme value distribution in fitting the extreme wind speeds data while also providing the best confidence bound for wind speed estimates with short return periods.

Dans cet article, les extrêmes des vitesses de vent sont étudiés dans la ville du Bénin au Nigéria. Les données concernent une période de 200 semaines. L'étude a été modélisée en utilisant une une famille de lois de probabilité dénommée Gumbel-Burr XII (GUBXII) qui a été introduite dans Osatohanmwen et al. (2017), comme étant un nouveau membre de la famille des distributions T-X, et ensuite avec la distribution généralisée des extrêmes. Une comparaison a été faite.

Citation

Download Citation

Patrick OSATOHANMWEN. Francis O OYEGUE. Sunday Martins OGBONMWAN. "Confidence Bounds for Extreme Wind Speed Estimates: A Comparison For the Gumbel - Burr XII Distribution and Classical Extreme Value Distributions." Afr. J. Appl. Stat. 4 (1) 259 - 271, January 2017. https://doi.org/10.16929/ajas/2017.259.214

Information

Published: January 2017
First available in Project Euclid: 16 May 2019

Digital Object Identifier: 10.16929/ajas/2017.259.214

Subjects:
Primary: 60-07 , 62F10 , 62G32

Keywords: AIC , confdence bound , Delta method , Extreme values , maximum likelihood , return period , wind speeds

Rights: Copyright © 2017 The Statistics and Probability African Society

Vol.4 • No. 1 • January 2017
Back to Top