A minimal characterization of the covariance matrix
暂无分享,去创建一个
SummaryLetX be ak-dimensional random vector with mean vectorμ and non-singular covariance matrix Σ. We show that among all pairs (a, Δ),a ∈ IRk, Δ ∈ IRk×k positive definite and 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 and scale.
[1] P. L. Davies,et al. Asymptotic behaviour of S-estimates of multivariate location parameters and dispersion matrices , 1987 .
[2] P. Rousseeuw. Multivariate estimation with high breakdown point , 1985 .
[3] Werner A. Stahel,et al. Robust Statistics: The Approach Based on Influence Functions , 1987 .