The distribution of minimizing maximum entropy: Alternative to Weibull distribution for wind speed


Shamilov A., USTA İ., Kantar Y. M.

WSEAS Transactions on Mathematics, vol.5, no.6, pp.695-700, 2006 (Scopus) identifier

  • Publication Type: Article / Article
  • Volume: 5 Issue: 6
  • Publication Date: 2006
  • Journal Name: WSEAS Transactions on Mathematics
  • Journal Indexes: Scopus
  • Page Numbers: pp.695-700
  • Keywords: Akaike information criteria, Bayes information criteria, Chi-square, Correlation coefficient, MaxEnt, MaxEnt distribution, MinMaxEnt distribution, Root mean square error, Wind speed data
  • Anadolu University Affiliated: Yes

Abstract

In the present study, we have constructed MinMaxEnt distribution alternative to the Weibull distribution for the wind speed. MinMaxEnt distribution minimizes the special maximum entropy functional in the concrete set of moment functions, therefore, maximizes the amount of information generated by the given random sample. MinMaxEnt distribution of wind speed data measured in Eskisehir is also obtained and compared with the Weibull distribution by using various statistical criterias. Consequently, it is determined that MinMaxEnt distributions is more convenient for wind speed distributions, compared to the Weibull which is widely used in wind studies.