Two types of recursive estimators are developed for the variance components 2 and 2 of the dynamic linear model: non-Bayesian and Bayesian. From a frequentist point of view, both types of estimators are mean square consistent. The non-Bayesian estimator of 2 is also unbiased.
Estimating variance components in linear models
β Scribed by Friedrich Pukelsheim
- Book ID
- 107880033
- Publisher
- Elsevier Science
- Year
- 1976
- Tongue
- English
- Weight
- 222 KB
- Volume
- 6
- Category
- Article
- ISSN
- 0047-259X
No coin nor oath required. For personal study only.
π SIMILAR VOLUMES
In mixed linear models with two variance components, classes of estimators improving on ANOVA estimators for the variance components and the ratio of variances are constructed on the basis of the invariant statistics. Out of the classes, consistent, improved and positive estimators are singled out.
The main purpose of this paper is to treat an unsolved problem discussed by S.K. Mitra (1977) concerning the solution of matrix equations which occur in the MINQUE theory of estimatin covariance components model of C.R. Rao (1972 3 covariance components in a . Canonical representation of a singular