The Adjustive Ridge Regression Estimator in a Linear Regression Model with Application to Portland cement Dataset

摘要

For the linear regression model y = Xβ +e with e～N(0, σ2In), the ordinary least squares estimator (OLSE) is not desirable any longer when the collinearity is present. In this case, various biased estimators were proposed to improve the OLSE in the literature. In the present paper, we propose a new estimator called to be adjustive ridge regression estimator and suggest viewing it as a good substitution for the OLSE in practical problems by illustrating the superiority via the dataset on Portland cement. Detailed discussions are given.