基于线性回归模型参数向量的先验信息提出一类新的s-K 估计改进s-K 估计,并在均方误差阵意义下,得到了这类估计分别优于最小二乘估计、广义岭估计、Stein估计及s-K估计的充要条件。
A new class of s-K estimators,called the modified s-K estimators was proposed based on the prior information for the parameter vector in a linear model. The necessary and sufficient conditions that the new estimators are respectively superior to the least squares estimator,ridge estimators,Stein estimators and s-K estimators were obtained.