ESTIMATING BOUNDED MEAN VECTOR IN MULTIVARIATE NORMAL: THE GEOMETRY OF HARTIGAN ESTIMATOR
Main Article Content
Abstract
The problem on estimating a multivariate normal mean Np(θ; I) when the vector mean is bounded awaked interest practical and theoretical. Under such hypothesis it's possible to obtain estimators which dominate the sample mean estimator in relation to square loss. Generalizing previous results obtained, for univariate normal, J.A. Hartigan obtained, for multivariate normal with independent components, a Bayes estimator defined on a bounded closed convex set, with non-empty interior, which dominates the sample mean estimator. In this work, this result is presented in details for the case where the restriction set is a sphere centered at origin. A geometrical interpretation, useful to understand the phenomenon, is presented. Others estimators based on Gatsonis et. al. (1987) are proposed and the risks of all these estimators are compared through simulations, for the cases of dimensions p = 1 and p = 2.
Article Details
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).