A Cramér-Rao type lower bound for estimators with values in a manifold

A Cramér-Rao type lower bound for minimum loss unbiased estimators with values in a manifold is derived, and the corresponding notion of efficiency is investigated. A by-product is a generalisation of the concept of covariance of a multivariate statistic to one of a statistic with values in a manifold.

MoreLess

Year of publication:	1991
Authors:	Hendriks, Harrie
Published in:	Journal of Multivariate Analysis. - Elsevier, ISSN 0047-259X. - Vol. 38.1991, 2, p. 245-261
Publisher:	Elsevier
Keywords:	Cramer-Rao inequality minimum variance unbiased estimation unbiased estimators with values in a manifold Hessian Fisher information covariance efficiency Weingarten map exponential family of probability distributions mean location Fisher-von Mises distributions integral manifold

More details

Type of publication:	Article
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005152893