Outlier detection for skewed data

  title={Outlier detection for skewed data},
  author={Mia Hubert and Stephan Van der Veeken},
Most outlier detection rules for multivariate data are based on the assumption of elliptical symmetry of the underlying distribution. We propose an outlier detection method which does not need the assumption of symmetry and does not rely on visual inspection. Our method is a generalization of the Stahel-Donoho outlyingness. The latter approach assigns to each observation a measure of outlyingness, which is obtained by projection pursuit techniques that only use univariate robust measures of… CONTINUE READING


Publications citing this paper.
Showing 1-10 of 39 extracted citations


Publications referenced by this paper.
Showing 1-10 of 32 references

Breakdown properties of multivariate location estimators

  • D. L. Donoho
  • Qualifying paper, Harvard University,
  • 1982
Highly Influential
10 Excerpts

Robuste Schätzungen: infinitesimale Optimalität und Schätzungen von Kovarianzmatrizen

  • W. A. Stahel
  • PhD thesis, ETH Zürich,
  • 1981
Highly Influential
10 Excerpts

The bagplot: A bivariate boxplot

  • P. J. Rousseeuw, I. Ruts, J. W. Tukey
  • The American Statistician,
  • 1999
Highly Influential
7 Excerpts

A robustification of Independent Component Analysis

  • G. Brys, M. Hubert, P. J. Rousseeuw
  • Journal of Chemometrics,
  • 2005
Highly Influential
3 Excerpts

A robust measure of skewness

  • G. Brys, M. Hubert, A. Struyf
  • Journal of Computational and Graphical Statistics…
  • 2004
Highly Influential
3 Excerpts

The behavior of the Stahel-Donoho robust multivariate estimator

  • R. A. Maronna, V. J. Yohai
  • Journal of the American Statistical Association,
  • 1995
Highly Influential
2 Excerpts

Similar Papers

Loading similar papers…