On ridge parameter estimators under stochastic subspace hypothesis

被引：10

作者：

Arashi, M. ^{[1
]}

Kibria, B. M. Golam ^{[2
]}

Valizadeh, T. ^{[3
]}

机构：

[1] Shahrood Univ Technol, Sch Math Sci, Dept Stat, Shahrood, Iran

[2] Florida Int Univ, Dept Math & Stat, Miami, FL 33193 USA

[3] Amirkabir Univ Technol, Sch Math & Comp Sci, Dept Stat, Tehran, Iran

来源：

JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION | 2017年 / 87卷 / 05期

关键词：

MSE; restricted estimator; ridge regression; simulation study; subspace hypothesis; REGRESSION; SIMULATION;

D O I：

10.1080/00949655.2016.1239104

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

This paper considers several estimators for estimating the restricted ridge parameter estimators. A simulation study has been conducted to compare the performance of these estimators. Based on the simulation study we found that, increasing the correlation between the independent variables has positive effect on the mean square error (MSE). However, increasing the value of. has negative effect on MSE. When the sample size increases, the MSE decreases even when the correlation between the independent variables is large. Two real life examples have been considered to illustrate the performance of the estimators.

引用

页码：966 / 983

页数：18

共 29 条

[1] Mean squared error matrix comparisons of some biased estimators in linear regression [J].