THE BIASED ESTIMATORS OF THE PARAMETERS IN THE TWO SEEMINGLY UNRELATED REGRESSION SYSTEM

For the two seemingly unrelated regression system （1.1）, some authors have investigated the improvement of the least squares estimators of the regression parameters. These improved estimators are all unbiased and have many advantages. But their mean square errors are large when the design matrix is ill-conditioned. In this paper, we proposed two kinds of biased estimators which have some superiority to the unbiased estimators.