Penerapan Regresi Ridge dalam Mengatasi Multikolinearitas pada Analisis Indeks Pembangunan Manusia di Sumatera Utara

Abstract

Multicollinearity problems are often found in linear regression analysis. Multicollinearity can be detected in several ways, including by calculating the Variance Inflation Factor (VIF) and Tolerance (TOL) values. To overcome the problem of multicollinearity, in this study used a method, namely the ridge regression method. This regression is a modification of the least squares method by adding a small bias constant c on the diagonal of the matrix. From the research results, the linear regression model obtained by the ridge regression method is