Bayesian statistical inference on elliptical matrix distributions

Kai Tai Fang; Runze Li

doi:10.1006/jmva.1998.1816

Bayesian statistical inference on elliptical matrix distributions

Kai Tai Fang^*, Runze Li

^*Corresponding author for this work

Research output: Contribution to journal › Journal article › peer-review

24 Citations (Scopus)

Abstract

In this paper we are concerned with Bayesian statistical inference for a class of elliptical distributions with parameters μ and Σ. Under a noninformative prior distribution, we obtain the posterior distribution, posterior mean, and generalized maximim likelihood estimators of μ and Σ. Under the entropy loss and quadratic loss, the best Bayesian estimators of Σ are derived as well. Some applications are given.

Original language	English
Pages (from-to)	66-85
Number of pages	20
Journal	Journal of Multivariate Analysis
Volume	70
Issue number	1
DOIs	https://doi.org/10.1006/jmva.1998.1816
Publication status	Published - Jul 1999
Externally published	Yes

Scopus Subject Areas

Statistics and Probability
Numerical Analysis
Statistics, Probability and Uncertainty

User-Defined Keywords

Elliptical matrix distributions
Entropy loss
Posterior mean
Quadratic loss

Access to Document

10.1006/jmva.1998.1816

Cite this

@article{f38b2fd3924848fbbde187b34ad2bc3c,

title = "Bayesian statistical inference on elliptical matrix distributions",

abstract = "In this paper we are concerned with Bayesian statistical inference for a class of elliptical distributions with parameters μ and Σ. Under a noninformative prior distribution, we obtain the posterior distribution, posterior mean, and generalized maximim likelihood estimators of μ and Σ. Under the entropy loss and quadratic loss, the best Bayesian estimators of Σ are derived as well. Some applications are given.",

keywords = "Elliptical matrix distributions, Entropy loss, Posterior mean, Quadratic loss",

author = "Fang, {Kai Tai} and Runze Li",

note = "Funding Information: This work was supported by a Hong Kong RGC grant, the Statistical Research and Consultancy Centre, Hong Kong Baptist University.",

year = "1999",

month = jul,

doi = "10.1006/jmva.1998.1816",

language = "English",

volume = "70",

pages = "66--85",

journal = "Journal of Multivariate Analysis",

issn = "0047-259X",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - Bayesian statistical inference on elliptical matrix distributions

AU - Fang, Kai Tai

AU - Li, Runze

N1 - Funding Information: This work was supported by a Hong Kong RGC grant, the Statistical Research and Consultancy Centre, Hong Kong Baptist University.

PY - 1999/7

Y1 - 1999/7

N2 - In this paper we are concerned with Bayesian statistical inference for a class of elliptical distributions with parameters μ and Σ. Under a noninformative prior distribution, we obtain the posterior distribution, posterior mean, and generalized maximim likelihood estimators of μ and Σ. Under the entropy loss and quadratic loss, the best Bayesian estimators of Σ are derived as well. Some applications are given.

AB - In this paper we are concerned with Bayesian statistical inference for a class of elliptical distributions with parameters μ and Σ. Under a noninformative prior distribution, we obtain the posterior distribution, posterior mean, and generalized maximim likelihood estimators of μ and Σ. Under the entropy loss and quadratic loss, the best Bayesian estimators of Σ are derived as well. Some applications are given.

KW - Elliptical matrix distributions

KW - Entropy loss

KW - Posterior mean

KW - Quadratic loss

UR - http://www.scopus.com/inward/record.url?scp=0033164043&partnerID=8YFLogxK

U2 - 10.1006/jmva.1998.1816

DO - 10.1006/jmva.1998.1816

M3 - Journal article

AN - SCOPUS:0033164043

SN - 0047-259X

VL - 70

SP - 66

EP - 85

JO - Journal of Multivariate Analysis

JF - Journal of Multivariate Analysis

IS - 1

ER -

Bayesian statistical inference on elliptical matrix distributions

Abstract

Scopus Subject Areas

User-Defined Keywords

Access to Document

Other files and links

Fingerprint

Cite this