Bayes estimators for linear models with less than full rank (Q3716043)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3945071
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Bayes estimators for linear models with less than full rank |
scientific article; zbMATH DE number 3945071 |
Statements
Bayes estimators for linear models with less than full rank (English)
0 references
1982
0 references
non full rank design matrix
0 references
estimability
0 references
relative merits
0 references
linear model
0 references
alternative forms
0 references
generalized ridge
0 references
generalized contraction estimators
0 references
average mean squared errors
0 references
0.8130282163619995
0 references
0.7974624037742615
0 references
0.7903881669044495
0 references
0.7868227958679199
0 references
The Bayes estimator for a parametric function of the coefficients of a linear model are considered. Five alternative forms of the estimator are derived when the design matrix does not have full rank. Some of the ridge, generalized ridge, contraction and generalized contraction estimators are shown to be special cases of the alternative forms we have derived. Necessary and sufficient conditions for the comparisons of the average mean squared errors of these estimators are derived.
0 references