Open AccessConfidence intervals for the difference between the coefficients of variation of Weibull distributions for analyzing wind speed dispersion
Author(s) -
Manussaya La-ongkaew,
Sa-Aat Niwitpong,
Suparat Niwitpong
Publication year - 2021
Publication title -
peerj
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.927
H-Index - 70
ISSN - 2167-8359
DOI - 10.7717/peerj.11676
Subject(s) - confidence interval , statistics , weibull distribution , credible interval , percentile , monte carlo method , robust confidence intervals , wind speed , wind power , cdf based nonparametric confidence interval , mathematics , coverage probability , bayesian probability , meteorology , engineering , physics , electrical engineering
Wind energy is an important renewable energy source for generating electricity that has the potential to replace fossil fuels. Herein, we propose confidence intervals for the difference between the coefficients of variation of Weibull distributions constructed using the concepts of the generalized confidence interval (GCI), Bayesian methods, the method of variance estimates recovery (MOVER) based on Hendricks and Robey’s confidence interval, a percentile bootstrap method, and a bootstrap method with standard errors. To analyze their performances, their coverage probabilities and expected lengths were evaluated via Monte Carlo simulation. The simulation results indicate that the coverage probabilities of GCI were greater than or sometimes close to the nominal confidence level. However, when the Weibull shape parameter was small, the Bayesian- highest posterior density interval was preferable. All of the proposed confidence intervals were applied to wind speed data measured at 90-meter wind energy potential stations at various regions in Thailand.