Abstract
The simultaneous estimation of the characteristic roots of the scale matrix of the multivariatet-model is considered. The improved estimation strategies are developed in the light of a quadratic loss function. It is demonstrated analytically and numerically that the class of proposed estimators outperforms the class of usual estimators in the sense of having smaller risk.
Similar content being viewed by others
References
Dey DK (1988) Simultaneous estimation of eigenvalues. Ann Inst Statist Math 40:137–147
Dey DK (1990) Estimation of scale parameters in mixture distributions. Canad J Statist 18:171–178
Joarder AH (1992) Estimation of the scale matrix of a multivariatet-model. Unpublished PhD thesis. University of Western Ontario Canada
Joarder AH, Ali MM (1992) On some generalized Wishart expectations. Commun Statist — Theor Meth 21:283–294
Leung PL (1992) Estimation of eigenvalues of the scale matrix of the multivariateF distribution. Commun Statist — Theor Meth 21:1845–1856
Singh RS (1991) James-Stein rule estimators in linear regression models with multivariatet-distributed error. Austr J Statist 33:145–158
Sutradhar BC, Ali MM (1986) Estimation of the parameters of a regression model with a multivariatet error variable. Commun Statist — Theor Meth 15:429–450
Sutradhar BC (1990) Discrimination of observations into one of twot populations. Biometrics, 46:827–835
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Joarder, A.H., Ahmed, S.E. Estimation of the characteristic roots of the scale matrix. Metrika 44, 259–267 (1996). https://doi.org/10.1007/BF02614070
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02614070