Bayes linear estimator in the general linear model

Bayes linear estimators are derived by minimizing the Bayes risk with respect to the squared loss function. Non-unbiased estimators such as an ordinary ridge estimator, a typical shrinkage estimator, and a fractional rank estimator are either Bayes linear estimators or their limit points. First, we prove that Bayes linear estimators are linearly complete, but not necessarily linearly sufficient. Second, we derive a necessary and sufficient condition under which two Bayes linear estimators coincide. Finally, we obtain an equivalence condition such that the equality between two residual sums of squares holds when Bayes linear estimators are considered.