The RLS Positive‐Part Stein Estimator
提出RLS正部斯坦估计量，证明其在加权二次损失下优于传统版本，并讨论其在经济和农业经济研究中的应用。
Abstract The RLS Stein‐rule estimator of the classical normal linear regression model is formed by taking a linear combination of the least squares and restricted least squares estimators. Using a simple analytical device, we prove that the convex combination known as the RLS positive‐part Stein estimator dominates the conventional version under weighted quadratic loss. Possible uses for the positive‐part estimator in economic and agricultural economic research are discussed.