Summary
Using a general result due to Perlman (1972), we show that no invariant unbiased estimator for variance components (not necessarily being quadratic) qualifies to admissibility under the mean squared error risk function. Uniformly better estimates are easily given by multiplication of an unbiased estimator with some constant less than but close to one. The optimal choice of such a multiplier is investigated and more specific results are derived for quadratic estimation.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Kleffe, J.(1980). On recent progress of MINQUE theory-non-negative estimation, consistency, asymptotic normality and explicite formulae. Math.Operationsforsch. Statist. Ser.Statistics, 11, 563–588.
Kleffe, J. and Rao, J.N.K. (1983). On the existence of asymptotically unbiased nonnegative quadratic estimates of variance components in ANOVA models. Technical Report, Carleton University, Ottawa Dept. of Mathematics and Statistics.
LaMotte, L.R.(1977). A canonical form for the general linear model. Ann. Statist. 5, 787–789.
LaMotte, L.R. (1979). On admissible invariant quadratic estimators of variance components, in: Variance Components and Animal Breeding. Proc. of a Conference in Honor of C.R. Henderson, L.Dale Vleck and S.R. Searle ed.
Ogasawara, T. and Takahashi, M.(1951). Independence of quadratic forms in normal systems. J.Sei.Hiroshima University, 15, 1–9.
Olkin, I. and Selliah, J.B. (1975). Estimating covariances in a multivariate normal distribution, Technical Report No. 100, Dept. of Statistics, Stanford University.
Patterson, H.D. and Thompson, R.(1975). Maximum likelihood estimation of components of variance. In: Proc. of 8th International Biometric Conference, 197–207.
Perlman, M.D. (1972). Reduced mean squared error estimation for several parameters. Sankhya Ser. B, 89–92.
Pukelsheim, F.(1981). On the existence of unbiased nonnegative estimates of variance components. Ann. Statist. 9, 293–299.
Pukelsheim, F. (1978). Examples for unbiased nonnegative estimation in variance component models. Technical Report No.113, Dept. of Statistics, Stanford University.
Rao, C.R.(1973). Linear Statistical Inference and its Applications. Second Edition. John Wiley, New York.
Seely, J.(1971). Quadratic subspaces and completeness. Ann.Math. Statist. 42, 710–721.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1985 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kleffe, J. (1985). Some Remarks on Improving Unbiased Estimators by Multiplication with a Constant. In: Caliński, T., Klonecki, W. (eds) Linear Statistical Inference. Lecture Notes in Statistics, vol 35. Springer, New York, NY. https://doi.org/10.1007/978-1-4615-7353-1_12
Download citation
DOI: https://doi.org/10.1007/978-1-4615-7353-1_12
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-96255-9
Online ISBN: 978-1-4615-7353-1
eBook Packages: Springer Book Archive