The multiColl package versus other existing packages in R to detect multicollinearity

Abstract

The main goal of this paper is to show the advantages of the multiColl package in R, comparing its results with other existing packages in R for the treatment of multicollinearity.

The main contributions of this paper can be summarized as follows: a) It was clarified that not all measures are adequate to detect all kinds of multicollinearity; b) When the model contains non-quantitative variables, unlike other packages existing in R analyzed in this paper, the multiColl package allows us to obviate this kind of variable; and c) The superiority of the multiColl package has been shown in comparison to other packages existing in R (except for the colldiag command of the perturb package) to detect the degree of multicollinearity existing in a simple linear regression.