A class of method of moments estimators for the two-parameter gamma family. | [Pakistan Journal of Statistics • 2003]

Author(s):

1. Douglas P. Wiens: Department of Mathematical & Statistical Sciences, University of Alberta, Canada

2. Julian Cheng: Department of Electrical and Computer Engineering, University of Alberta, Canada

3. Norman C. Beaulieu: Department of Electrical and Computer Engineering, University of Alberta, Canada

Abstract:

The authors have introduced and studied a class of estimators, for the two-parameter Gamma family, which are based on three moments of the sample. The members of this family are easy to computer, relative to the maximum likelihood estimator or its commonly used approximation. A limiting member of the family is surprisingly efficient. The other members of the class, while approaching in efficiency this limiting member, possess a property of robustness against “inliers” observations that are aberrant by virtue of being too small.

Page(s): 129-141

DOI: DOI not available

Published: Journal: Pakistan Journal of Statistics, Volume: 19, Issue: 1, Year: 2003

Keywords:

Keywords are not available for this article.

References:

References are not available for this document.

Citations

Citations are not available for this document.

Citations

Downloads

Views