Almost unbiased Liu-type estimator for Tobit regression and its application

Abstract

This paper introduces a new estimator for the Tobit regression model when the multicollinearity exists. This estimator is called almost unbiased Tobit Liu-Type estimator, which is obtained by a mixture between the Liu-Type estimator and almost unbiased estimator. This estimator avoids the negative effects of multicollinearity on the maximum likelihood estimator. Furthermore, we check the superiority of the new estimator over the maximum likelihood estimator, Liu-Type estimator and almost unbiased estimator according to the simulated mean square error SMSE criteria. In addition, we run the simulation study to investigate the effects of a group of factors on the performance of the new estimator. Finally, to check the benefits of the new estimator via the real data, we use two applications for the Tobit regression, the English Premier League data and Egyptian agricultural GDP data.