Summary: One of the problems encountered in linear regression models is called multicollinearity problem which is an approximately linear relationship between the explanatory variables. This problem causes the estimated parameter values to be highly sensitive to small changes in the data. In order to reduce the impact of this problem on the model parameters, alternative biased estimators to the ordinary least squares estimator have been proposed in the literature. In this study, we propose a new biased estimator that can be an alternative to existing estimators. The superiority of this estimator over other biased estimators is analyzed in terms of matrix mean squared error. In addition, two different Monte Carlo simulation experiments are carried out to examine the performance of the biased estimators under consideration. A numerical example is given to evaluate the performance of the proposed estimator against other biased estimators.