A New Estimator for the Gaussian Linear Regression Model with Multicollinearity

In this study, a new two-parameter biased estimator for estimating the parameter of a multicollinearity linear regression model is developed. We compared the performance of the proposed estimator with some existing estimators in terms of the mean squared error matrix. The theoretical comparisons and simulation results revealed that the proposed two-parameter estimator dominate other estimators under certain conditions using the mean squared error. The simulation result shows that the ordinary least squares estimator has the lowest performance. Among the one-parameter estimator, the ridge regression estimator dominates the Liu estimator. However, the proposed method dominates both the one-parameter and two-parameter estimators. We also discuss the estimations of the biasing parameters. A famous real-life dataset is analyzed to support both theoretical and simulation results.