Hoerl and Kennard (1970) suggested the ridge regression estimator as an alternative to the Ordinary Least Squares (OLS) estimator in the presence of multicollinearity. This article proposes new methods for estimating the ridge parameter in case of ordinary ridge regression. A simulation study evaluates the performance of the proposed estimators based on the Mean Squared Error (MSE) criterion and indicates that, under certain conditions, the proposed estimators perform well compared to the OLS estimator and another well-known estimator reviewed.
Khalaf, Ghadban and Iguernane, Mohamed
"Ridge Regression and Ill-Conditioning,"
Journal of Modern Applied Statistical Methods:
2, Article 18.
Available at: http://digitalcommons.wayne.edu/jmasm/vol13/iss2/18