Modified Jackknife Estimator in Linear Regression Model

Abstract

In multiply linear regression model, more method was proposed to improve the biased estimation. One of this method   jackknife biased estimation is considered one of important methods to address the high variance and multicollinearity problems. In this paper, we propose biased estimator called modified jackknife estimator (MJE) based on the jackknife Liu type estimator (JLTE).  We derive the MJE estimator as the solution of the following problem, minimize β'βsubiect to . In section 4, we conduct a simulation study based on the mean squares error (MSE) to assess how well the new estimator performs in comparison to a few jackknife biased estimators.  We demonstrate that, in comparison to several other jackknife estimators, the MJE has good qualities. Finally, using data from real-world situations, we demonstrate how well this estimator performs.