×

Projection-pursuit approach to robust dispersion matrices and principal components: Primary theory and Monte Carlo. (English) Zbl 0595.62060

Robust estimators of covariance matrices and principal components based on the projection-pursuit technique are proposed. Rotational equivariance is proved. The level of the breakdown point is discussed. Qualitative robustness and consistency are shown for elliptic distributions.
A Monte Carlo study illustrates the method and indicates that the method has a high breakdown point even for asymmetric contamination.
Reviewer: J.Á.Višek

MSC:

62H99 Multivariate analysis
62F35 Robustness and adaptive procedures (parametric inference)
62H25 Factor analysis and principal components; correspondence analysis
62H12 Estimation in multivariate analysis
PDFBibTeX XMLCite
Full Text: DOI