Robustifying principal component analysis with spatial sign vectors

Statistics and Probability Letters

Volumn 82, Issue 4, 2012, Pages 765-774

  Taskinen, Sara ^a,b   Koch, Inge ^c   Oja, Hannu ^d  

^a UNIVERSITY OF JYVÄSKYLÄ   (Finland)

^b UNIVERSITY OF NEW SOUTH WALES   (Australia)

^c UNIVERSITY OF ADELAIDE   (Australia)

^d UNIVERSITY OF TAMPERE   (Finland)

Author keywords

Affine equivariance; Efficiency; Influence function; Robustness; Spatial sign vector

Indexed keywords

EID: 84856045016     PISSN: 01677152     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.spl.2012.01.001     Document Type: Article

References (24)

1. Anderson T.W. Asymptotic theory for principal component analysis. Ann. Math. Statist. 1963, 34:122-148.
2. Anderson T.W. An Introduction to Multivariate Statistical Analysis 1984, Wiley, New York. second ed.
3. Croux C., Haesbroeck G. Principal component analysis based on robust estimators of the covariance or correlation matrix: influence functions and efficiencies. Biometrika 2000, 87:603-618.
4. Croux C., Ollila E., Oja H. Sign and rank covariance matrices: statistical properties and application to principal components analysis. Statistical Data Analysis Based on the L1-Norm and Related Methods 2002, 257-271. Birkhauser, Basel. Y. Dodge (Ed.).
5. Devlin S.J., Gnanadesikan R., Kettenring J.R. Robust estimation of dispersion matrices and principal components. J. Amer. Statist. Assoc. 1981, 76:354-362.
6. Dümbgen L. On Tyler's M-functional of scatter in high dimension. Ann. Inst. Statist. Math. 1998, 50:471-491.
7. Dümbgen L., Tyler D. On the breakdown properties of some multivariate M-functionals. Scandinavian J. Statist. 2005, 32:247-264.
8. Frahm G. Asymptotic distributions of robust shape matrices and scales. J. Multivariate Anal. 2009, 100:1329-1337.
9. Gervini D. Robust functional estimation using the spatial median and spherical principal components. Biometrika 2008, 95:587-600.
10. Hampel F.R., Ronchetti E.M., Rousseeuw P.J., Stahel W.J. Robust Statistics: The Approach Based on Influence Functions 1986, Wiley, New York.
11. Hettmansperger T.P., Randles R.H. A practical affine equivariant multivariate median. Biometrika 2003, 89:851-860.
12. Locantore N., Marron J.S., Simpson D.G., Tripoli N., Zhang J.T., Cohen K.L. Robust principal components for functional data. TEST 1999, 8:1-73.
13. Marden J.I. Some robust estimates of principal components. Statist. Probab. Lett. 1999, 43:349-359.
14. Nordhausen K., Oja H. Multivariate L1 methods: the package MNM. J. Stat. Softw. 2011, 43:1-28.
15. Oja H. Multivariate Nonparametric Methods with R: An Approach Based on Spatial Signs and Ranks 2010, Springer, New York.
16. Oja H., Randles R. Multivariate nonparametric tests. Statist. Sci. 2004, 19:598-605.
17. Paindaveine D. A canonical definition of shape. Statist. Probab. Lett. 2008, 78:2240-2247.
18. Sirkiä S., Taskinen S., Oja H. Symmetrised M-estimators of scatter. J. Multivariate Anal. 2007, 98:1611-1629.
19. Tanaka Y. Sensitivity analysis in principal component analysis: influence on the subspace spanned by principal component. Commun. Stat. 1988, 17:3157-3175.
20. Taskinen S., Croux C., Kankainen A., Ollila E., Oja H. Influence functions and efficiencies of the canonical correlation and vector estimates based on scatter and shape matrices. J. Multivariate Anal. 2006, 97:359-384.
21. Tyler D.E. A distribution-free M-estimator of multivariate scatter. Ann. Statist. 1987, 15:234-251.
22. Tyler D.E. A note on multivariate location and scatter statistics for sparse data sets. Statist. Probab. Lett. 2010, 80:1409-1413.
23. Vardi Y., Zhang C.-H. A modified Weiszfeld algorithm for the Fermat-Weber location problem. Math. Program. 2001, 90:559-566.
24. Visuri S., Koivunen V., Oja H. Sign and rank covariance matrices. J. Statist. Plann. Inference 2000, 91:557-575.