scientific article; zbMATH DE number 4028602

From MaRDI portal
Publication:3769755





[For the entire collection see Zbl 0583.00027.] This paper deals with the linear model \(\mu =EY=X\beta\), Cov y\(=\sigma^ 2V\) and the question is investigated when a homogeneous linear estimator L'Y is an admissible estimator of C'\(\mu\) with respect to the Loewner ordering of the usual risk function. The Stein paradox and the Cohen result on linear admissible estimators are mentioned. They are obtained for \(X=I_ n\) and are generalized here to arbitrary X. The admissible linear estimators of C'\(\mu\) are characterized by the admissible estimators of \(\mu\). Also, linear admissible estimators of \(\beta\) are considered in the full rank case.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3769755)