The compression LS estimate of regression coefficient in multivariate linear model

DOI10.1007/BF02463717MaRDI QIDQ1333840zbMATH OpenFDO

Authors Shiji Chen, Zhi-Bin Zeng

Publication date 6 February 1995

Published in Applied Mathematics and Mechanics. (English Edition) (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/BF02463717

zbMATH Keywords

compression least squares estimationleast squares estimatormean square errormultivariate linear modelnumerical stabilityregression coefficientsrelative efficiencyridge estimationtuning numbers

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Linear regression; mixed models (62J05) Ridge regression; shrinkage estimators (Lasso) (62J07)

Review text

Compression least squares estimation of the regression coefficients is considered when the design matrix is near singularity in a multivariate linear model. The mean square error of the estimator is shown to be less than the mean square error of the least squares estimator of the regression coefficients by choosing proper tuning numbers. Besides, admissibility, numerical stability and relative efficiency of the compression least squares estimate are investigated, and a method of selecting the tuning number for practical use is suggested.

Recommendations

• scientific article; zbMATH DE number 4082772
• scientific article; zbMATH DE number 221947
• scientific article; zbMATH DE number 5047351
• Linear regression.
• Admissibility and improvement of universal least squares estimators

Cites work

• scientific article; zbMATH DE number 3812757 (Why is no real title available?)
• scientific article; zbMATH DE number 53802 (Why is no real title available?)
• scientific article; zbMATH DE number 123423 (Why is no real title available?)
• The inefficiency of least squares

Cited in

(1)

• Economical method for estimation of nonzero coefficients of linear normal regression

This page was built for publication: The compression LS estimate of regression coefficient in multivariate linear model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1333840)