Multicollinearity and parameter estimates in a linear model.

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

The effects of multicollinearity on a value of a particular parameter estimate obtained in the calibration of a linear regression model is explicitly demonstrated. Assessment of the effects of multicollinearity from the magnitude of the correlation coefficient between two independent variables is shown to be inadequate. The extent of errors in parameter estimation due to model misspecification are also discussed. -Author

Original languageEnglish (US)
Pages (from-to)64-71
Number of pages8
JournalGeographical Analysis
Volume14
Issue number1
StatePublished - 1982
Externally publishedYes

Fingerprint

linear model
calibration
regression
Values
effect
parameter
parameter estimation

ASJC Scopus subject areas

  • Geography, Planning and Development

Cite this

Multicollinearity and parameter estimates in a linear model. / Fotheringham, Stewart.

In: Geographical Analysis, Vol. 14, No. 1, 1982, p. 64-71.

Research output: Contribution to journalArticle

@article{1077f0964b99499c8ebb6059c90e722f,
title = "Multicollinearity and parameter estimates in a linear model.",
abstract = "The effects of multicollinearity on a value of a particular parameter estimate obtained in the calibration of a linear regression model is explicitly demonstrated. Assessment of the effects of multicollinearity from the magnitude of the correlation coefficient between two independent variables is shown to be inadequate. The extent of errors in parameter estimation due to model misspecification are also discussed. -Author",
author = "Stewart Fotheringham",
year = "1982",
language = "English (US)",
volume = "14",
pages = "64--71",
journal = "Geographical Analysis",
issn = "0016-7363",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - Multicollinearity and parameter estimates in a linear model.

AU - Fotheringham, Stewart

PY - 1982

Y1 - 1982

N2 - The effects of multicollinearity on a value of a particular parameter estimate obtained in the calibration of a linear regression model is explicitly demonstrated. Assessment of the effects of multicollinearity from the magnitude of the correlation coefficient between two independent variables is shown to be inadequate. The extent of errors in parameter estimation due to model misspecification are also discussed. -Author

AB - The effects of multicollinearity on a value of a particular parameter estimate obtained in the calibration of a linear regression model is explicitly demonstrated. Assessment of the effects of multicollinearity from the magnitude of the correlation coefficient between two independent variables is shown to be inadequate. The extent of errors in parameter estimation due to model misspecification are also discussed. -Author

UR - http://www.scopus.com/inward/record.url?scp=0019897532&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0019897532&partnerID=8YFLogxK

M3 - Article

VL - 14

SP - 64

EP - 71

JO - Geographical Analysis

JF - Geographical Analysis

SN - 0016-7363

IS - 1

ER -