Recovering Jackknife Ridge Regression Estimates from OLS Results

Abstract

The aim of this paper is addressing or recalculate the estimation methods in multiple linear regression model when there is a problem of Multicollinearity in this model like the ridge regression for Hoerl and Kannard, Baldwin estimator (HKB) and Jackknifed ridge regression estimator (JRR) using least-squares estimators which the last are the best unbiased estimators, consistent and linear. In this paper we proposed a formula to calculate the above estimators easily depending on the least-squares estimator, this treatment as a mathematical formula faster than the HKB estimator that depend on reducing the variance and JRR estimator that depend on reducing the bias. We used numerical examples of the pricing method in comprehensive quality and environmental quality as air pollution in places as pricing environment. After the comparison JRR and HKB estimates are superior to the OLS estimates under the mean squared error (MSE) criterion. 2000 Mathematics Subject Classification: Primary 62J07.