Stein-rule restricted regression estimator in a linear regression model with nonspherical disturbances

In the present paper, we propose a Stein-rule estimator for the general linear regression model with nonspherical disturbances and a set of linear restrictions binding the regression coefficients. The asymptotic risk properties of the proposed estimator under a quadratic loss structure are derived, and a sufficient condition For the proposed estimator to dominate the feasible generalized restricted least squares estimator in large samples is presented. The small sample behavior of the proposed estimator is studied via a Monte-Carlo experiment.