摘要
将根方估计作一拓广,提出了回归系数的广义根方估计,证明了广义根方估计能更有效地改善最小二乘估计,并给出了广义根方估计的显示解和确定偏参数的方法。
The generalized root power estimation of regression coefficient is proposed on the basis of the broadening of root power estimation.It is proved that the generalized root power estimation may effectively improve the least square estimation.Its explicit solution and the method to evaluate the optimal biased parameter are given
出处
《西安矿业学院学报》
北大核心
1996年第3期275-279,共5页
Journal of Xi'an University of Science & Technology
关键词
多元线性模型
广义根方估计
回归系数
multivariate linear model least square estimation generalized root power estimation admissibility mean square error