LEAST COST FORMULATIONS, LTD.
824 Timberlake Drive, Virginia Beach, VA 23464-3239
Tel: (757) 467-0954 Fax: (757) 467-2947
E-mail:
URL:
/
TECHNICAL REPORT
NUMBER:
TR323
DATE: 2012 August 13
TITLE:
Statistical analysis of interlaboratory studies. XXXVI. When robust statistics
make sense in proficiency test studies.
AUTHOR:
R. A. LaBudde
ABSTRACT: The use of ‘robust’ statistical estimators for measures of central location and
variation are discussed. Robust estimators for measures of central location are
non-controversial and acceptable. Robust estimators for measures of variation,
such as reproducibility standard deviation, are heavily biased and therefore
deprecated. Examples of performance of robust estimators are given for four
example distributions (normal, lognormal, student-t and gamma).
KEYWORDS:
1) PT
2) REPRODUCIBILITY 3) VARIANCE
4) ROBUST 5) OUTLIER
6) IQR
REL.DOC.:
TR322
REVISED:
Copyright 2012 by Least Cost Formulations, Ltd.
All Rights Reserved
Recommended to OMB by Committee on Statistics: 07-17-2013
Reviewed and approved by OMB: 07-18-2013
20
51
