A Comparison of Methods for the Detection of Outliers in Multivariate Data

  • Hadi, Ali-S. (Department of Statistics, Comell University) ;
  • Joo, Hye-Seon (Department of Statistics, Comell University) ;
  • Son, Mun-S. (Department of Mathematics and Statistics, 16 Colchester Avenue, University of Vermont)
  • Published : 1996.08.01

Abstract

Numerous classical as well as robust methods have been proposed in the literature for the detection of multiple outlier in multivariate data. The effectiveness and power of each of these methods have not been thoroughly investigated. In this paper we first reduce the vast number of outlier detection methods to a small number of viable ones. This reduction is based on previous work of other researches and on some theoretical arguments. Then we design and implement a Monte Carlo experiment for comparing these methods. The main goal of our study is to determine which methods are most powerful in the detection of multiple outlier and in dealing with the masking and swamping problems. The results of the Monte Carlo study indicate that two of the methods seem to hace better performances than the others for the detection of multiple outlier in multivariate data.

Keywords

References

  1. Journal of the Royal Statistical. Society(C) v.36 no.2 A New Graphical Method for Detecting Single and Multiple Outliers in Univariate ad Nultivariate Data Bacon-Shone, J.;Fung, W. K.
  2. Outiers in Statistical data(2nd edition) Barnett, V.;Lewis, T.
  3. Technometrics v.25 outlier...s Beckman, R.;Cook, R. D.
  4. Applied Statistics v.29 Robust Procedures in Multivariate Analysis Ⅰ: Robust Covariance Estimation Campbell, N. A.
  5. Applied Statistics v.41 Sequential Application of Wilks's Multivariate Outlier Test Caroni, C.;Prescott, P.
  6. Multivariate Behavioral Research v.20 A Method for Recovering Outliers to Improve Factor Analytic Results Comrey, A. L.
  7. Annals of Statistics v.15 no.3 Asymptotic Behavior of S-estimates of Multivariate Location Prameters and Dispersion Matrices Davis, P. L.
  8. Journal of the Statistical Computation and Simulation v.30 no.1 Critical Values for Testing in Multivarate statistical Outliers Fung, W. K.
  9. Proceedings of the Busivess and Economic Statistics Section Influential Observations in Data Analysis Gasko, M.;Donoho, D. L.
  10. Journal of the Royal Statistical. Society, series(B) v.54 Identifying Multiple Outliers in Multivariate Data Hadi, A. S.
  11. Journal of the Royal Statistical. Society, series(B) v.56 no.2 A Modification of a Method for the Detection of Ouliers in Multivariate Samples Hadi, A. S.
  12. Journal of the American Statistical Association v.88 no.414 Procedures for the Identification of Multiple Outliers in Linear Models Hadi, A. S.;Simonoff, J. S.
  13. Robust Statistics: The Approach Based on Influence Functions Hampel, F. R.;Ronchetti, E. M.;Rousseeuw, P. J.;Stahel, W. A.
  14. Identification of Outliers Hawkins, D. M.
  15. Applied Statistics v.17 Healy, M. J. R.
  16. Robust Statistics Huber, P.
  17. Annals of Statistics v.17 no.4 On the Relation Between S-Estimators and M-Estimators of Multivariate Location and Covariance Lopuhaa, H. P.
  18. Biometrica v.57 Measures of Multivariate Skewness and Kurtosis with Application Mardia, K. V.
  19. Sankhya v.B36 Application of Some Measures of Multivariate Skewness and Kurtosis in Testin Normality and Robusteness Studies Mardia, K. V.
  20. Annals of Statistics v.4 no.1 robust M-Estimators of Multivariate Location and Scatter Moronna, R. A.
  21. Nultivariate Behavioral Research v.23 Evaluating Outlier Identification Test: Mahalanobis D Squared and Comrey Dk Rasmussen, J. L.
  22. Biometrics v.31 Generalization of the Grap Test for Detection of multivariate outliers Rohlf, F. J.
  23. Mathematical Statistics and Applications v.B Multivariate Estimation with High Breakdown point Rousseeuw, P. J.;W. Grossmann(ed.);G. Pflug(ed.);I. Vincze(ed.);W. Wertz(ed.)
  24. Roubst Regression and Outlier Detection Rousseeuw, P. J.;Leroy, A. M.
  25. Journal of the American Statistical Association v.85 Unmasking Multivariate Outliers and Leverage Points (with discussion) Rousseeuw, P. J.;Zomeren, B. C.
  26. Directions in Robust Statistics and Diagnostics Robust Distances: simulations and Cuiff Values Rousseeuw, P. J.;Zomeren, B. C.;W. stahel(ed.);S. Weisberg(ed.)
  27. Robust and Nonlinear Time series Analysis, Lecture Notes in Statistics Robust Regression by Means of S Estimators Rousseeuw, P. J.;Yohai, V. J.
  28. Annals of Statistics v.10 no.3 Detection of Multivariate Normal Outliers Schwager, S. J.;Margolin, B. H.
  29. Directions in Robust Statistics and Diagnostics General Approaches to Stepwise Identification of Unusual Values in Data Analysis Simonoff, J. S.;W. Stahel(ed.);S.weisberg(ed.)
  30. Sakhya v.A25 Multivariate Statistical otliers Wilks, S. S.