摘要
对线性模型中回归系数的Stein型估计,双h类估计和双l类估计作了改进,使得它们在设计阵的任何情形下都能一致优于最小二乘估计。
The purpose of this paper is to modify three calsses of biased estimations (Stein-typed estimations, double-h ridged estimations, double-l ridged estimations) so that these improved estimations could always be better than the least squared estimation in the sense of MSE.
关键词
最小二乘估计
线性模型
有偏估计
least squared estimation, Stein-typed estimations, Double-h-ridged estimations, double-l ridged estimations, mean square error
