On the cumulants of affine equivariant estimators in elliptical families

Rudolf Grübel, David M Rocke

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Given a statistical model for data which take values in Rd and have elliptically distributed errors, and affine equivariant estimators μ̂ and μ̂ of a mean vector in Rd⊗Rn and a d × d scatter matrix, expressions are given for the covarances of the estimators in terms of their expectations and some unknown constants that depend on the model and the estimator. Higher order cumulants are also developed. These results place considerable constraints on the possible cumulants of μ̂ and μ̂, as wel as those of estimators of higher order behavior such as multivariate skewness and kurtosis. These expressions are obtained using tensor methods.

Original languageEnglish (US)
Pages (from-to)203-222
Number of pages20
JournalJournal of Multivariate Analysis
Volume35
Issue number2
DOIs
StatePublished - 1990

Fingerprint

Equivariant Estimator
Cumulants
Tensors
Estimator
Higher Order
Kurtosis
Skewness
Scatter
Statistical Model
Tensor
Unknown
Family
Statistical Models
Model

Keywords

  • maximum likelihood
  • multivariate location
  • multivariate regression
  • robust estimation
  • seemingly unrelated regression
  • tensor methods

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Numerical Analysis
  • Statistics and Probability

Cite this

On the cumulants of affine equivariant estimators in elliptical families. / Grübel, Rudolf; Rocke, David M.

In: Journal of Multivariate Analysis, Vol. 35, No. 2, 1990, p. 203-222.

Research output: Contribution to journalArticle

@article{29b4ec98ebc7426795edf363b63b6a1c,
title = "On the cumulants of affine equivariant estimators in elliptical families",
abstract = "Given a statistical model for data which take values in Rd and have elliptically distributed errors, and affine equivariant estimators μ̂ and μ̂ of a mean vector in Rd⊗Rn and a d × d scatter matrix, expressions are given for the covarances of the estimators in terms of their expectations and some unknown constants that depend on the model and the estimator. Higher order cumulants are also developed. These results place considerable constraints on the possible cumulants of μ̂ and μ̂, as wel as those of estimators of higher order behavior such as multivariate skewness and kurtosis. These expressions are obtained using tensor methods.",
keywords = "maximum likelihood, multivariate location, multivariate regression, robust estimation, seemingly unrelated regression, tensor methods",
author = "Rudolf Gr{\"u}bel and Rocke, {David M}",
year = "1990",
doi = "10.1016/0047-259X(90)90025-D",
language = "English (US)",
volume = "35",
pages = "203--222",
journal = "Journal of Multivariate Analysis",
issn = "0047-259X",
publisher = "Academic Press Inc.",
number = "2",

}

TY - JOUR

T1 - On the cumulants of affine equivariant estimators in elliptical families

AU - Grübel, Rudolf

AU - Rocke, David M

PY - 1990

Y1 - 1990

N2 - Given a statistical model for data which take values in Rd and have elliptically distributed errors, and affine equivariant estimators μ̂ and μ̂ of a mean vector in Rd⊗Rn and a d × d scatter matrix, expressions are given for the covarances of the estimators in terms of their expectations and some unknown constants that depend on the model and the estimator. Higher order cumulants are also developed. These results place considerable constraints on the possible cumulants of μ̂ and μ̂, as wel as those of estimators of higher order behavior such as multivariate skewness and kurtosis. These expressions are obtained using tensor methods.

AB - Given a statistical model for data which take values in Rd and have elliptically distributed errors, and affine equivariant estimators μ̂ and μ̂ of a mean vector in Rd⊗Rn and a d × d scatter matrix, expressions are given for the covarances of the estimators in terms of their expectations and some unknown constants that depend on the model and the estimator. Higher order cumulants are also developed. These results place considerable constraints on the possible cumulants of μ̂ and μ̂, as wel as those of estimators of higher order behavior such as multivariate skewness and kurtosis. These expressions are obtained using tensor methods.

KW - maximum likelihood

KW - multivariate location

KW - multivariate regression

KW - robust estimation

KW - seemingly unrelated regression

KW - tensor methods

UR - http://www.scopus.com/inward/record.url?scp=38249017907&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38249017907&partnerID=8YFLogxK

U2 - 10.1016/0047-259X(90)90025-D

DO - 10.1016/0047-259X(90)90025-D

M3 - Article

AN - SCOPUS:38249017907

VL - 35

SP - 203

EP - 222

JO - Journal of Multivariate Analysis

JF - Journal of Multivariate Analysis

SN - 0047-259X

IS - 2

ER -