The Hájek convolution theorem and Pitman closeness for regular Bayes estimators

Rok, strany: 1999, 27 - 35

For possibly vector valued parameters, the Hájek (1970) [J. Hájek: A characterization of limiting distributions of regular estimates, Z. Wahrsch. verw. Geb. 14 (1970), 323–330] convolution theorem provides asymptotic optimality properties of regular estimators under quite general regularity assumptions, and this basic result has been extended to more diverse setups during the past three decades. Allowing shrinking prior distributions, regular Bayes estimators are considered, and an analogous convolution theorem is incorporated in the study of (generalized) Pitman closeness of regular Bayes estimators. Related isomorphism results for asymptotically normal family of estimators, arising in robust statistical inference, are also appraised in the current context.

ISO 690:

Sen, P. 1999. The Hájek convolution theorem and Pitman closeness for regular Bayes estimators. In Tatra Mountains Mathematical Publications, vol. 17, no.3, pp. 27-35. 1210-3195.

APA:

Sen, P. (1999). The Hájek convolution theorem and Pitman closeness for regular Bayes estimators. Tatra Mountains Mathematical Publications, 17(3), 27-35. 1210-3195.