Another Cautionary Note About Multicollinearity Problem In A Linear Regression: A New Look At An Old Issue
Ijomah Maxwell Azubuike
Department of Mathematics/Statistics, University of Port Harcourt, Choba, Rivers State
Nduka Ethelbert Chinaka
Department of Mathematics/Statistics, University of Port Harcourt, Choba, Rivers State
Keywords: multicollinearity, collinearity, regressors, regressand, correlation
Abstract
The problem of Multicollinearity in a regression model is often considered as the correlation between two or more explanatory variables (regressors). Though the literature on ways of coping with collinearity is extensive, relatively little effort has been made to clarify the conditions under which collinearity affect estimates developed with multiple regression analysis or how pronounced those effects are. In this paper, effort is made to capture the nature of collinearity within the regressors as well as between the regressand and regressors in both near perfect positive and negative correlation. Furthermore, most literature on how number of correlated regressors affect the degree of collinearity in a regression model are very scanty. Our results provide critical insight that both helps avoid misleading interpretations and yields better understanding for the impact of intercorrelation among predictor variables in linear regression analyses.