Parameter estimation using different information-theoretical divergence measures

Shamilov A., MERT KANTAR Y., USTA İ.

WSEAS Transactions on Information Science and Applications, vol.3, no.12, pp.2355-2359, 2006 (Scopus) identifier

  • Publication Type: Article / Article
  • Volume: 3 Issue: 12
  • Publication Date: 2006
  • Journal Name: WSEAS Transactions on Information Science and Applications
  • Journal Indexes: Scopus
  • Page Numbers: pp.2355-2359
  • Keywords: Downhill simplex method, J measure, Kullback-Leibler measure, Parameter estimation methods, Weibull distribution
  • Anadolu University Affiliated: Yes

Abstract

In this paper, a approach for parameter estimation is presented by using different divergence information measures. A comparison of the proposed approach and usual methods, such as the moment method, the least square method, is also given. The presented optimization approach is applied to estimate two-parameter Weibull distribution which is widely used in reliability, life testing, survival analysis, engineering and wind energy studies. Also, analysis of validation of the a approach used the different information-theoretical divergence measures on real wind data is presented.