You are not currently logged in.
Access JSTOR through your library or other institution:
Empirical Bayes Estimators of Normal Covariance Matrix
Divakar Sharma and K. Krishnamoorthy
Sankhyā: The Indian Journal of Statistics, Series A (1961-2002)
Vol. 47, No. 2 (Jun., 1985), pp. 247-254
Published by: Indian Statistical Institute
Stable URL: http://www.jstor.org/stable/25050540
Page Count: 8
Preview not available
Let S follow a Wishart distribution, with n degrees of freedom and parameter Σ. Haff (1979b, 1980) has proposed estimators of Σ which dominate the best multiples of S under certain loss functions. We consider some other loss functions, show that these estimators are empirical Bayes and compare one of them with the best multiples of S.
Sankhyā: The Indian Journal of Statistics, Series A (1961-2002) © 1985 Indian Statistical Institute