ridgregextra: an alternative R package for ridge regression parameter estimation

This article presents an R package ridgregextra which provides various outputs related to ridge regression analysis. The package automatically finds the ridge parameter value k which makes the variance inflation factors (VIF) values closest to 1 while keeping them above 1 as addressed in M.H. Kutner et al. [Applied Linear Statistical Models, 5th ed., McGraw Hill, New York, 2004] opposite to other R packages available for ridge regression. This automatic and principled approach for selecting the ridge parameter enhances model interpretability and stability, addressing a key gap in existing softwares. Besides, most of the other packages also do not suggest a ridge parameter value k which can be confusing for practitioners. Additionally, there is no method in the literature which produces the VIF-based ridge parameter estimator. Ridgregextra package provides ridge regression results depending on the automatically selected ridge parameter value while giving regression tables (VIF, mean squared error (MSE), coefficient of determination (R2), regression coefficients (Betas), standard errors of the coefficients (Stdbetas) for the ridge parameter values generated between certain lower and upper bounds defined by the user. In addition, it provides three sets of graphs consisting k vs. VIF values, Betas and the Stdbetas. These visual diagnostics facilitate a comprehensive understanding of the trade-offs involved in ridge regularization, empowering users to make informed decisions.