Journal of the Korean Statistical Society

The Korean Statistical Society (한국통계학회)
권호 표지

Resistant Singular Value Decomposition and Its Statistical Applications

  • Park, Yong-Seok (Department of Statistics, Research Institute of Information and Communication, Pusan National University, Pusan, 609-735, Korea.) ;
  • Huh, Myung-Hoe
  • Published : 1996.03.01
Download PDF
⟨ Previous Next ⟩

Abstract

The singular value decomposition is one of the most useful methods in the area of matrix computation. It gives dimension reduction which is the centeral idea in many multivariate analyses. But this method is not resistant, i.e., it is very sensitive to small changes in the input data. In this article, we derive the resistant version of singular value decomposition for principal component analysis. And we give its statistical applications to biplot which is similar to principal component analysis in aspects of the dimension reduction of an n x p data matrix. Therefore, we derive the resistant principal component analysis and biplot based on the resistant singular value decomposition. They provide graphical multivariate data analyses relatively little influenced by outlying observations.

Keywords

References

  1. Regression Diagnostics: Identifying Influential Data and Sources of Collinearity Belsley, D.;Kuh, E.;Welsch, R. E.
  2. Technometrics v.20 The biplot as a diagnostic tool for models of two-way tables Bradu, D.;Gabriel, K. R.
  3. Applied Statistics v.29 Robust procedures in multivariate analysis Ⅰ: robust covariance estimation Campbell, N. A.
  4. Unpublished Ph.D. dissertation, University of Korea Resistant Principal Component Analysis, Biplot And Correspondence Analysis Choi, Y. S.
  5. Residuals and Influence in Regression Cook, R. D.;Weisberg, S.
  6. Biometrika v.62 Robust estimation and outlier detection with correlation coefficients Devlin, S. J.;Gnanadesikan, R.;Kettenring, J. R.
  7. Graphical Exploratory Data Analysis du Toit, S. H. C.;Steyn, A. G. W.;Stumpf, R. H.
  8. Biometrika v.58 The biplot graphics display of matrices with applications to principal component analysis Gabriel, K. R.
  9. Technometrics v.21 Lower rank approximation of matrices by least squares with any choice of weights Gabriel, K. R.;Zamirs
  10. Biometrics v.28 Robust estimates, residuals, and outlier detection with multireponse data Gnanadesikan, R.;Kettenring, J. R.
  11. Theory and Applications of Correspondence Analysis Greenacre, M. J.
  12. Understanding Robust and Exploratory Data Analysis Hoaglin, D. C.;Mosteller, F.;Tukey, J. W.
  13. Robust Statistics Huber, P. J.
  14. Principal Component Analysis Jolliffe, I. T.
  15. Applied Multivariate Statistical Analysis Johnson, R. A.;Wichern, D. W.
  16. Multivariate Descriptive Statistical Analysis: Correspondence Analysis and Related Techniques for Large Matrices Lebart, L.;Morineau, A.;Warwick, K.
  17. Exploring Data Tables, Trends and Shapes Robust Regression Li, G.;Hoaglin, D. C.(ed.);Mosteller, F.(ed.);Tukey, J. W.(ed.)
  18. Annals of Statistics v.4 Robust M-estimators of multivariate location and scatter Maronna, R. A.
  19. Applied Statistics v.33 Robust methods in the assessment of multivariate normality Mattews, J. N. S.
  20. Multivariate Data Analysis Murtagh, F.;Heck, A.
  21. Rev Statistique Appliquee v.36 N'analyse en composantes principales robuste Rivest, L. P.;Plante, N.
  22. Multivariate Observations Seber, G. A.