Remedy of multicollinearity using Ridge regression

One of the main problems in the model is two or more of the explanatory variables in the sample are overlapping, and this overlapping indicates a multicollinearity problem.

Ridge regression is one of the famous methods for remedy of the multicollinearity problem because it enables us to keep these explanatory variables, which violate the assumption of independency in the model. In this paper, we used real data and R package to find the multicollinearity problem in the established linear model between two explanatory variables. One of the multicollinearity problems solutions is to omit the explanatory variables, which cause the multicollinearity. We show that by using the ridge regression, we get the new estimates of the new model without omitting any of the explanatory variables.

Keywords: Ridge regression – multicollinearity - ordinary least square - singular value      decomposition - generalized cross validation - Lawless and Wang method.