REDUCING THE MAXIMUM RISK OF REGRESSION-ESTIMATORS BY POLYHEDRAL PROJECTION

被引：3
|
作者：
SCHMIDT, K [1 ]
STAHLECKER, P [1 ]
机构：
[1] UNIV HAMBURG,INST STAT & OKONOMETRIE,HAMBURG,GERMANY
来源：
JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION | 1995年 / 52卷 / 01期
关键词：
LINEAR REGRESSION; INEQUALITY RESTRICTIONS; MINIMAX ESTIMATION; INEQUALITY RESTRICTED LEAST SQUARES ESTIMATOR; PROJECTION METHODS;
D O I：
10.1080/00949659508811648
中图分类号：
TP39 [计算机的应用];
学科分类号：
081203 ; 0835 ;
摘要：
We consider the linear regression model where the parameter space is restricted by linear inequalities. The effect of projecting estimators that may violate this polyhedral prior information is analyzed. We show that projected estimators dominate their unprojected counterparts. The magnitude of the performance gain is investigated in a Monte Carlo Study.
引用
收藏
页码：1 / 15
页数：15
相关论文
共 50 条
12345