A new biased estimation method in tobit regression: theory and application

Abstract

In this study, the effects of multicollinearity on the maximum likelihood estimator are analyzed in the tobit regression model. It is known that the near-linear dependencies in the design matrix affect the maximum likelihood estimation negatively, namely, the standard errors become so large so that the estimations are said to be inconsistent. Therefore, a new biased estimator being a generalization of the well-known Liu estimator is introduced as an alternative to the maximum likelihood estimator. Mean squared error properties of the estimators are investigated theoretically. In order to evaluate the performances of the estimators, a Monte Carlo simulation study is designed and simulated mean squared error is used as a performance criterion. Finally, the benefits of the new estimator is illustrated via real data applications.