A New Measure of Predictor Variable Importance in Multiple Regression

Open Access

1 August 1978

journal article
other
Published by SAGE Publications in Journal of Marketing Research

Vol. 15 (3) , 356-360
https://doi.org/10.1177/002224377801500305

Abstract

Ambiguity surrounds any importance measure in cases in which predictor variables are correlated. However, a new measure is proposed that has attractive properties, such as providing individual contributions that are both non-negative and sum to R². The new measure is compared with four other commonly used measures and its advantages over each of them are pointed out.

Keywords

IMPORTANCE MEASURE
VARIABLE IMPORTANCE
PREDICTOR VARIABLES
NEW MEASURE
INDIVIDUAL CONTRIBUTIONS
ATTRACTIVE PROPERTIES
PROVIDING INDIVIDUAL
AMBIGUITY SURROUNDS
USED MEASURES
REGRESSION AMBIGUITY

This publication has 3 references indexed in Scilit:

Multiple regression in psychological research and practice.
Psychological Bulletin, 1968
The Minimal Transformation to Orthonormality
Psychometrika, 1966
Orthogonal Predictors: A Possible Resolution of the Hoffman-Ward Controversy
Psychological Reports, 1962