New algorithms for M-estimation of multivariate scatter and location

DOI10.1016/J.JMVA.2015.11.009MaRDI QIDQ901290zbMATHOpenAlexWikidataFDO

Authors Lutz Dümbgen, Klaus Nordhausen, Heike Schuhmacher

Publication date 23 December 2015

Published in Journal of Multivariate Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1312.6489

zbMATH Keywords

Newton-Raphson algorithmmatrix exponential functionTaylor expansionfixed-point algorithm

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12)

Abstract: We present new algorithms for $M$-estimators of multivariate scatter and location and for symmetrized $M$-estimators of multivariate scatter. The new algorithms are considerably faster than currently used fixed-point and related algorithms. The main idea is to utilize a second order Taylor expansion of the target functional and to devise a partial Newton-Raphson procedure. In connection with symmetrized $M$-estimators we work with incomplete $U$-statistics to accelerate our procedures initially.

Recommendations

• Some Results on the Existence, Uniqueness, and Computation of the M-Estimates of Multivariate Location and Scatter
• On the computation of symmetrized M-estimators of scatter
• \(M\)-functionals of multivariate scatter
• A distribution-free M-estimator of multivariate scatter
• The DetS and DetMM estimators for multivariate location and scatter

Cites work

• scientific article; zbMATH DE number 3678955 (Why is no real title available?)
• A cautionary note on robust covariance plug-in methods
• A curious likelihood identity for the multivariate t-distribution
• A distribution-free M-estimator of multivariate scatter
• Convergence Behavior of the em algorithm for the multivariate t -distribution
• Differentiability of \(t\)-functionals of location and scatter
• Invariant co-ordinate selection (with discussion)
• Mixture Densities, Maximum Likelihood and the EM Algorithm
• On Tyler's \(M\)-functional of scatter in high dimension
• Redescending \(M\)-estimates of multivariate location and scatter
• Robust Statistics
• Statistical Analysis of Financial Data in S-Plus
• Symmetrised M-estimators of multivariate scatter
• \(M\)-functionals of multivariate scatter

Cited in

(17)

• Non-Gaussian component analysis: testing the dimension of the signal subspace
• A robust deterministic affine-equivariant algorithm for multivariate location and scatter
• Inference and mixture modeling with the elliptical Gamma distribution
• Multivariate τ-Estimators for Location and Scatter
• \(M\)-functionals of multivariate scatter
• M-estimation with incomplete and dependent multivariate data
• Robust sparse covariance estimation by thresholding Tyler's M-estimator
• A review of Tyler's shape matrix and its extensions
• Robustly fitting Gaussian graphical models -- the R package robFitConGraph
• On the computation of symmetrized M-estimators of scatter
• The DetS and DetMM estimators for multivariate location and scatter
• The flood algorithm -- a multivariate, self-organizing-map-based, robust location and covariance estimator
• Some Results on the Existence, Uniqueness, and Computation of the M-Estimates of Multivariate Location and Scatter
• Symmetrised M-estimators of multivariate scatter
• Gini covariance matrix and its affine equivariant version
• Tyler's and Maronna's M-estimators: non-asymptotic concentration results
• Approximating symmetrized estimators of scatter via balanced incomplete \(U\)-statistics

Describes a project that uses

Uses Software

• R
• MASS (R)
• ICSNP
• ICS
• fastM

This page was built for publication: New algorithms for \(M\)-estimation of multivariate scatter and location

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q901290)