A Monte Carlo Study of Recent Ridge Parameters

A number of procedures have been developed for finding biased estimators of regression parameters. One of these procedures is the ridge regression. In this paper, a new approach to obtain the ridge parameter (K) is suggested and then evaluated by Monte Carlo simulations. A large number of different models were investigated, where the number of observations, the strength of correlation between the explanatory variables and the distribution of the error terms have been varied. The mean squared of error (MSE) is used as criterion to examine the performance of the proposed estimators when compared with other well-known estimators. Under certain conditions, it is shown that at least one of the proposed estimators have a smaller (MSE) than the ordinary least squared estimator (OLS) and Hoerl and Kennard (1970a) estimator (HK).