Abstract
It is shown that in a general statistical decision problem Bayes procedures under convex loss can possess the following property: The induced prior measure of the convex hull of the range of the corresponding Bayes rule is equal to zero. A stronger statement holds in the case of binomial distributions.
Similar content being viewed by others
References
Brown, L., Cohen, A., and Strawderman, W. E. (1976). A complete class theorem for strict monotone ratio with applications.Ann. Statist. 4, 712–722.
Fu, J. (1985). An exponential rate of likelihood ratio estimators for location parameters.Statist. and Prob. Lett. 3, 351–356.
Karlin, S., and Shapley, L. S. (1953).Geometry of Moment Spaces. Memorandum of the Americal Mathematical Society, No. 12. American Mathematical Society, Providence.
Karlin, S., and Studden, W. J. (1966).Tchebycheff Systems with Applications in Analysis and Statistics. Interscience, New York.
Rockafellar, R. T. (1970).Convex Analysis. Princeton University Press, Princeton.
Rojo, J. (1988). Proper priors yielding linear Bayes estimators for the natural parameter of an exponential family. Manuscript.
Skibinsky, M. (1986). Principal representations and canonical moment sequences for distribution on an interval.J. Math. Anal. Appl. 120, 95–118.
Skibinsky, M., and Rukhin, A. L. (1989). Admissible estimators of binomial probability and the inverse Bayes rule map.Ann. Inst. Statist. Math. 41, 699–716.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Rukhin, A.L., Skibinsky, M. Bayes estimators whose range has convex hull of zero prior probability. J Theor Probab 4, 465–474 (1991). https://doi.org/10.1007/BF01258749
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF01258749