A simple motivation for James-Stein estimators
A purely frequentist development of James-Stein shrinkage estimators of the multivariate normal mean under quadratic loss functions is presented, which allows for an intuitive interpretation of these estimators as best estimators of best linear 'estimators' of the mean vector.
Year of publication: |
1991
|
---|---|
Authors: | Gupta, Arjun K. ; Peña, Edsel A. |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 12.1991, 4, p. 337-340
|
Publisher: |
Elsevier |
Keywords: | Admissibility multivariate normal quadratic loss risk function shrinkage estimator |
Saved in:
Saved in favorites
Similar items by person
-
Bayes estimation for the Marshall-Olkin exponential distribution
Peña, Edsel A., (1990)
-
Compound p-value statistics for multiple testing procedures
Habiger, Joshua D., (2014)
-
Information‐theoretic model‐averaged benchmark dose analysis in environmental risk assessment
Piegorsch, Walter W., (2013)
- More ...