Abstract
This article mainly discusses the admissibility of quadratic estimate of covariance in pseudo-elliptical distribution. Under the quadratic loss function, the necessary and sufficient conditions that a quadratic estimator is an admissible estimator of covariance in the class of quadratic estimators are obtained. A complete class of the quadratic estimator class is also given.
Similar content being viewed by others
References
P. Cheng, Q. G. Wu, and G. Y. Li, Admissibility of quadratic estimates of 2-order moment about the origin, Acta Mathematicae Applicatae Sinica, 1983, 6(1): 18–28.
Q. G. Wu, P. Cheng, and G. Y. Li, Admissibility of the quadratic form estimates of error variance in linear models, Scientia Sinica, 1981, 7: 815–825.
Q. G. Wu, P. Cheng, and G. Y. Li, Further discussion on admissibility of the quadratic form estimates of error variance in linear models, Journal of Systems Science and Mathematical Sciences, 1981, 1(2): 112–127.
X. Z. Xu, A necessary and sufficient condition that the quadratic estimate for error variance in linear model is D-admissible, Acta Mathematicae Applicatae Sinica, 1992, 15(3): 410–427.
X. Z. Xu, Admissibility of quadratic estimator of error variance in normal linear model, Acta Mathematica Sinica, 1996, 39(5): 609–618.
X. Z. Xu and Q. G. Wu, The admissible quadratic estimators of covariance, Acta Math. Appl. Sinica, 2000, 23(1): 141–150.
Author information
Authors and Affiliations
Additional information
The research was partially supported by the Doctoral Programme of Higher Education (No. 20020027010) of China.
Rights and permissions
About this article
Cite this article
Cui, H., Gao, X. Quadratic Admissible Estimate of Covariance in Pseudo-Elliptical Contoured Distribution. Jrl Syst Sci & Complex 19, 236–255 (2006). https://doi.org/10.1007/s11424-006-0236-8
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/s11424-006-0236-8