Background Image
Table of Contents Table of Contents
Previous Page  175 / 274 Next Page
Information
Show Menu
Previous Page 175 / 274 Next Page
Page Background

LEAST COST FORMULATIONS, LTD.

824 Timberlake Drive, Virginia Beach, VA 23464-3239

Tel: (757) 467-0954 Fax: (757) 467-2947

E-mail:

office@lcftld.com

URL:

http://lcfltd.com/

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

122