Robust statistical analysis of interlaboratory studies

Research output: Contribution to journalArticle

45 Citations (Scopus)

Abstract

A common procedure in testing analytical methods is to send a portion of each of a number of samples to each of several laboratories. The results of such a study are submitted to statistical analysis to determine the two important variance components in the problem: replication error and laboratory bias. Outliers are relatively common in these data both among laboratory effects and among the residuals. This paper presents a method of analysis for interlaboratory studies that is robust to the existence of outliers and long-tailed distributions of random effects. Theoretical considerations as well as a Monte Carlo study are adduced as support for this new technique.

Original languageEnglish (US)
Pages (from-to)421-431
Number of pages11
JournalBiometrika
Volume70
Issue number2
DOIs
StatePublished - Aug 1983

Fingerprint

Outlier
Statistical Analysis
Statistical methods
statistical analysis
Variance Components
Monte Carlo Study
Random Effects
Analytical Methods
Replication
Monte Carlo method
Testing
analytical methods
methodology
Statistical analysis
testing
sampling
Outliers
Random effects
Monte Carlo study
Analytical methods

Keywords

  • Biweight estimate
  • Huber estimate
  • Least squares
  • Monte Carlo method
  • Outlier
  • Random effects
  • Robust estimation
  • Variance component

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Applied Mathematics
  • Mathematics(all)
  • Statistics and Probability
  • Agricultural and Biological Sciences (miscellaneous)
  • Agricultural and Biological Sciences(all)

Cite this

Robust statistical analysis of interlaboratory studies. / Rocke, David M.

In: Biometrika, Vol. 70, No. 2, 08.1983, p. 421-431.

Research output: Contribution to journalArticle

@article{e7d998c547a1414facee4fcfa4fdc0d9,
title = "Robust statistical analysis of interlaboratory studies",
abstract = "A common procedure in testing analytical methods is to send a portion of each of a number of samples to each of several laboratories. The results of such a study are submitted to statistical analysis to determine the two important variance components in the problem: replication error and laboratory bias. Outliers are relatively common in these data both among laboratory effects and among the residuals. This paper presents a method of analysis for interlaboratory studies that is robust to the existence of outliers and long-tailed distributions of random effects. Theoretical considerations as well as a Monte Carlo study are adduced as support for this new technique.",
keywords = "Biweight estimate, Huber estimate, Least squares, Monte Carlo method, Outlier, Random effects, Robust estimation, Variance component",
author = "Rocke, {David M}",
year = "1983",
month = "8",
doi = "10.1093/biomet/70.2.421",
language = "English (US)",
volume = "70",
pages = "421--431",
journal = "Biometrika",
issn = "0006-3444",
publisher = "Oxford University Press",
number = "2",

}

TY - JOUR

T1 - Robust statistical analysis of interlaboratory studies

AU - Rocke, David M

PY - 1983/8

Y1 - 1983/8

N2 - A common procedure in testing analytical methods is to send a portion of each of a number of samples to each of several laboratories. The results of such a study are submitted to statistical analysis to determine the two important variance components in the problem: replication error and laboratory bias. Outliers are relatively common in these data both among laboratory effects and among the residuals. This paper presents a method of analysis for interlaboratory studies that is robust to the existence of outliers and long-tailed distributions of random effects. Theoretical considerations as well as a Monte Carlo study are adduced as support for this new technique.

AB - A common procedure in testing analytical methods is to send a portion of each of a number of samples to each of several laboratories. The results of such a study are submitted to statistical analysis to determine the two important variance components in the problem: replication error and laboratory bias. Outliers are relatively common in these data both among laboratory effects and among the residuals. This paper presents a method of analysis for interlaboratory studies that is robust to the existence of outliers and long-tailed distributions of random effects. Theoretical considerations as well as a Monte Carlo study are adduced as support for this new technique.

KW - Biweight estimate

KW - Huber estimate

KW - Least squares

KW - Monte Carlo method

KW - Outlier

KW - Random effects

KW - Robust estimation

KW - Variance component

UR - http://www.scopus.com/inward/record.url?scp=0000750772&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0000750772&partnerID=8YFLogxK

U2 - 10.1093/biomet/70.2.421

DO - 10.1093/biomet/70.2.421

M3 - Article

AN - SCOPUS:0000750772

VL - 70

SP - 421

EP - 431

JO - Biometrika

JF - Biometrika

SN - 0006-3444

IS - 2

ER -