An Inequality for a Measure of Deviation in Linear Models

Thomas Mathew, Kenneth Nordström

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

A matrix inequality is established that provides an upper bound for a quadratic form that involves the difference between two linear unbiased estimators of the same linear parametric function in a general linear model. Various special cases of the inequality are discussed. Certain inequalities that arise in the problem of outlier detection and prediction of observations come out as special cases. In addition, some extensions of Samuelson's inequality are also obtained.

Original languageEnglish
Pages (from-to)344-349
Number of pages6
JournalAmerican Statistician
Volume51
Issue number4
Publication statusPublished - Nov 1997
Externally publishedYes

Fingerprint

Linear Model
Deviation
Outlier Detection
Unbiased estimator
Quadratic form
Matrix Inequality
Upper bound
Prediction
Observation

Keywords

  • Outliers
  • Samuelson's inequality
  • Studentized residuals

ASJC Scopus subject areas

  • Mathematics(all)
  • Statistics and Probability

Cite this

Mathew, T., & Nordström, K. (1997). An Inequality for a Measure of Deviation in Linear Models. American Statistician, 51(4), 344-349.

An Inequality for a Measure of Deviation in Linear Models. / Mathew, Thomas; Nordström, Kenneth.

In: American Statistician, Vol. 51, No. 4, 11.1997, p. 344-349.

Research output: Contribution to journalArticle

Mathew, T & Nordström, K 1997, 'An Inequality for a Measure of Deviation in Linear Models', American Statistician, vol. 51, no. 4, pp. 344-349.
Mathew, Thomas ; Nordström, Kenneth. / An Inequality for a Measure of Deviation in Linear Models. In: American Statistician. 1997 ; Vol. 51, No. 4. pp. 344-349.
@article{35bbde9e2aa9460f8a222dd73cc85969,
title = "An Inequality for a Measure of Deviation in Linear Models",
abstract = "A matrix inequality is established that provides an upper bound for a quadratic form that involves the difference between two linear unbiased estimators of the same linear parametric function in a general linear model. Various special cases of the inequality are discussed. Certain inequalities that arise in the problem of outlier detection and prediction of observations come out as special cases. In addition, some extensions of Samuelson's inequality are also obtained.",
keywords = "Outliers, Samuelson's inequality, Studentized residuals",
author = "Thomas Mathew and Kenneth Nordstr{\"o}m",
year = "1997",
month = "11",
language = "English",
volume = "51",
pages = "344--349",
journal = "American Statistician",
issn = "0003-1305",
publisher = "American Statistical Association",
number = "4",

}

TY - JOUR

T1 - An Inequality for a Measure of Deviation in Linear Models

AU - Mathew, Thomas

AU - Nordström, Kenneth

PY - 1997/11

Y1 - 1997/11

N2 - A matrix inequality is established that provides an upper bound for a quadratic form that involves the difference between two linear unbiased estimators of the same linear parametric function in a general linear model. Various special cases of the inequality are discussed. Certain inequalities that arise in the problem of outlier detection and prediction of observations come out as special cases. In addition, some extensions of Samuelson's inequality are also obtained.

AB - A matrix inequality is established that provides an upper bound for a quadratic form that involves the difference between two linear unbiased estimators of the same linear parametric function in a general linear model. Various special cases of the inequality are discussed. Certain inequalities that arise in the problem of outlier detection and prediction of observations come out as special cases. In addition, some extensions of Samuelson's inequality are also obtained.

KW - Outliers

KW - Samuelson's inequality

KW - Studentized residuals

UR - http://www.scopus.com/inward/record.url?scp=0031500488&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031500488&partnerID=8YFLogxK

M3 - Article

VL - 51

SP - 344

EP - 349

JO - American Statistician

JF - American Statistician

SN - 0003-1305

IS - 4

ER -