Abstract
In this paper Γ-minimax estimation of a multivariate normal mean under arbitrary squared error loss is considered where the covariance matrix of the normal distribution is a known symmetric and positive definite matrix with unknown multiple. The set Γ is fixed by imposing restrictions on the vector of first moments and on the matrix of second moments as well as on the first moment of the unknown factor determining of the covariance matrix. Necessary and sufficient conditions are derived which ensure that an estimator is Γ-minimax and that a prior is least favorable in Γ.
Similar content being viewed by others
References
Berger, J.O. Statistical Decision Theory. 2d, ed., Springer, Berlin-Heidelberg-New York, 1985.
Casella, G. and Strawderman, W. Estimating a Bounded Normal Mean.Ann. Statist., 1981, 9: 868–876.
Chen, L. Gamma-minimax Estimators for a Multivariate Normal Mean.Chin. Ann. of Math., 1991, 12B(1): 45–57.
De Groot, M.H. Optimal Statistical Decisions. MeGraw-Hill, New York, 1970.
Jackson, D.A., O'Donovan, T.M., Zimmer, W.J. and Deely, J.J.G 2-minimax Estimators in the Exponential Family.Biometrika, 1970, 57: 439–443.
Kuks, J. An Approximately Minimax Estimator of the Mean of an One-dimensioal Normal Distribution.Izw. Acad. Nauk ESSR, 1983, 32: 268–273 (in Russian).
Ortega, J.M. and Rheinboldt, W.C. Iterative Solution of Nonlinear Equations in Several Variables. Academic Press, Orlando, 1970.
Raubut, B., Schmitz, N. and Zachow, E.W. Spieltheorie. Teubner, Stuttgart, 1979.
Solomon D.L. Λ-minimax Estimation of a Multivariate Location Parameter.J. Amer. Stat. Assoc., 1972, 67: 641–646.
Zinzius, E. Minimaxschätzer für den Mittelwertϑ einer normalverteilten Zufallsgröße mit bekannter Varianz bei vorgegebener oberer und unterer Schranke fürϑ.Math. Operationsforsch. Statist. (Ser. Statist.), 1981, 12: 551–557.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Chen, L., Lehn, J. Gamma-minimax estimators for the mean of a multivariate normal distribution with partially unknown covariance matrix. Acta Mathematicae Applicatae Sinica 11, 11–16 (1995). https://doi.org/10.1007/BF02012618
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02012618