first
prev
toc
next
last
Download Citation
Recommend
Mark
Alert
Link
Print
Admissible improvements of the least squares estimator *
Author:
Kurt Hoffmann a
| Affiliation: | a Sektion Mathematik, Humboldt-Universit t, Berlin, Postfach |
DOI:
10.1080/02331888008801547
Publication Frequency:
6 issues per year
Subjects:
Mathematical Statistics;
Statistical Theory & Methods;
Statistics;
Statistics for the Biological Sciences;
Stochastic Models & Processes;
Formats available:
PDF
(English)
Article Requests:
Order Reprints
:
Request Permissions
-
Single Article Purchase:
US$37.00
-
buy now
-
Sign In
- Online Sample
add to cart 
[ show other buying options ]
View Article:
View Article (PDF)
Abstract
A unified theory of linear estimation for the regression parameter is represented in eases where the regression parameter varies only in an arbitrary ellipsoid ≼N. Referring to the basic concepts of universal ≼N-admissibility and universal ≼N-improvements we derive universal ≼N-improvements of the least squares estimator being universally Gtf- admissible.
|
|
*
1A first version of this paper was gives at the 4th International Summer School on Problems of Model Choice and Parameter Estimation in Regression analysis, M
 hlhausen(GDR), May 1979.
|
| view references (16) |
|
Bookmark with:
|
CiteULike
Del.icio.us
BibSonomy
Connotea
More bookmarks
