Unbiasing procedures for the Weibull distribution? Be careful

Author

Montanari, G.C. ; Mazzanti, G. ; Cacciari, M. ; Fothergill, J.C.

Author_Institution

Istituto di Elettrotecnica Ind., Bologna Univ., Italy

fYear

1995

fDate

10-13 Jul 1995

Firstpage

175

Lastpage

179

Abstract

In this paper, unbiasing procedures for the maximum likelihood method applied to the two-parameter Weibull function are dealt with. The performance of unbiasing methods applied to expected values and point estimates of the Weibull parameters, as well as the way to use the Monte Carlo method for the estimation of the expected values, are discussed. It is shown that the accuracy of the unbiasing methods can be significantly affected by several factors, such as the value of the shape parameter and the estimation of the expected value, and that some methods can be successfully applied to the point estimates of the Weibull parameters

Keywords

Monte Carlo methods; Weibull distribution; electric breakdown; maximum likelihood estimation; Monte Carlo method; Weibull distribution; maximum likelihood method; point estimates; two-parameter Weibull function; unbiasing procedures; Distributed computing; Maximum likelihood estimation; Parameter estimation; Random number generation; Sampling methods; Shape; Weibull distribution;

fLanguage

English

Publisher

ieee

Conference_Titel

Conduction and Breakdown in Solid Dielectrics, 1995. ICSD'95., Proceedings of the 1995 IEEE 5th International Conference on

Conference_Location

Leicester

Print_ISBN

0-7803-2040-9

Type

conf

DOI

10.1109/ICSD.1995.522972

Filename

522972