A minimal characterization of the covariance matrix

作者: R. Grübel

DOI: 10.1007/BF02613285

关键词:

摘要: LetX be ak-dimensional random vector with mean vectorμ and non-singular covariance matrix Σ. We show that among all pairs (a, Δ),a ∈ IR k , Δ k×k positive definite symmetric andE(X−a)′ Δ−1(X−a)=k, (μ, Σ) is the unique pair which minimizes det Δ. This motivates certain robust estimators of location scale.

参考文章(3)
Elvezio M. Ronchetti, Peter J. Rousseeuw, Werner A. Stahel, Frank R. Hampel, Robust statistics: the approach based on influence functions ,(1986)