Shrunken Principal Component Least Squares Estimator

被引:0
|
作者
Sheng Zining [1 ]
机构
[1] Shanghai Maritime Univ, Dept Math, Shanghai 201612, Peoples R China
关键词
Multicollinearity; Shrunken Principal Component Estimator; Admissibility; Mean-squared Error; RIDGE-REGRESSION;
D O I
暂无
中图分类号
F3 [农业经济];
学科分类号
0202 ; 020205 ; 1203 ;
摘要
In this paper, we propose a Shrunken Principal Component Least Squares Estimator for the regression coefficient. Using one biased parameter, we can give different shrunken ratio to different eigenvector corresponding to the different eigenvalue. The admissibility of the estimator is discussed. And its mean-squared error is compared analytically with the Principal Component version. Finally, the performance of the new estimator is evaluated for a real data set.
引用
收藏
页码:1022 / 1025
页数:4
相关论文
共 50 条