Abstract:
We investigate the performance of robust estimates of multivariate location under nonstandard data contamination models such as componentwise outliers (i.e., contamination in each variable is independent from the other variables). This model brings up a possible new source of statistical error that we call "propagation of outliers." This source of error is unusual in the sense that it is generated by the data processing itself and takes place after the data has been collected. We define and derive the influence function of robust multivariate location estimates under flexible contamination models and use it to investigate the effect of propagation of outliers. Furthermore, we show that standard high-breakdown affine equivariant estimators propagate outliers and therefore show poor breakdown behavior under componentwise contamination when the dimension d is high. © Institute of Mathematical Statistics, 2009.
Registro:
Documento: |
Artículo
|
Título: | Propagation of outliers in multivariate data |
Autor: | Alqallaf, F.; Van Aelst, S.; Yohai, V.J.; Zamar, R.H. |
Filiación: | Kuwait University, Kuwait Ghent University, Belgium University of Buenos Aires, Argentina University of British Columbia, Canada Department of Statistics and Operations Research, Faculty of Science, Kuwait University, P.O. Box 5969, Safat-13060, Kuwait Department of Mathematics, University of Buenos Aires, Ciudad Universitaria, Pabellón 1, 1426 Buenos Aires, Argentina Department of Applied Mathematics and Computer Science, Ghent University, Krijgslaan 281 S9, B-9000 Gent, Belgium Department of Statistics, University of British Columbia, 6356 Agricultural Road, Vancouver, BC V6T 1Z2, Canada
|
Palabras clave: | Breakdown point; Contamination model; Independent contamination; Influence function; Robustness |
Año: | 2009
|
Volumen: | 37
|
Número: | 1
|
Página de inicio: | 311
|
Página de fin: | 331
|
DOI: |
http://dx.doi.org/10.1214/07-AOS588 |
Título revista: | Annals of Statistics
|
Título revista abreviado: | Ann. Stat.
|
ISSN: | 00905364
|
Registro: | https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_00905364_v37_n1_p311_Alqallaf |
Referencias:
- ALQALLAF, F., VAN AELST, S., YOHAI, V. J. and ZAMAR, R. H. (2006). A model for contamination in multivariate data. Technical report, Dept. Statistics, Univ. British Columbia, Vancouver. Available online at http://users.ugent.be/~svaelst/publications.html; BARNET, V. and LEWIS, T. (1994). Outliers in Statistical Data. Wiley, New York. MR1272911; CROUX, C., FILZMOSER, P., PISON, G., ROUSSEEUW, P.J., Fitting multiplicative models by robust alternating regressions (2003) Statist. Comput, 13, pp. 23-36. , MR1973864
- DAVIES, P.L., Asymptotic behavior of S-estimates of multivariate location parameters and dispersion matrices (1987) Ann. Statist, 15, pp. 1269-1292. , MR0902258
- DONOHO, D.L., (1982) Breakdown properties of multivariate location estimators. Qualifying paper, , Harvard Univ
- HAMPEL, F. R., RONCHETTI, E. M., ROUSSEEUW, P. J. and STAHEL, W. A. (1986). Robust Statistics: The Approach Based on Influence Functions. Wiley, New York. MR0829458; HE, X., SIMPSON, D.G., PORTNOY, S., Breakdown robustness of tests (1990) J. Amer. Statist. Assoc, 85, pp. 446-452. , MR1141746
- HE, X., SIMPSON, D.G., Lower bounds for contamination bias: Globally minimax versus locally linear estimation (1993) Ann. Statist, 21, pp. 314-337. , MR1212179
- HUBER, P.J., Robust estimation of a location parameter (1964) Ann. Math. Statist, 35, pp. 73-101. , MR0161415
- KENT, J.T., TYLER, D.E., Constrained M-estimation for multivariate location and scatter (1996) Ann. Statist, 24, pp. 1346-1370. , MR1401854
- LIU, L., HAWKINS, D.M., GHOSH, S., YOUNG, S.S., Robust singular value decomposition analysis of microarray data (2003) Proc. Natl. Acad. Sci. USA, 100, pp. 13167-13172. , MR2016727
- LOPUHAÄ, H.P., On the relation between S-estimators and M-estimators of multivariate location and covariance (1989) Ann. Statist, 17, pp. 1662-1683. , MR1026304
- LOPUHAÄ, H.P., Multivariate ô -estimators for location and scatter (1991) Canad. J. Statist, 19, pp. 307-321. , MR1144148
- LOPUHAÄ, H.P., ROUSSEEUW, P.J., Breakdown points of affine equivariant estimators of multivariate location and covariance matrices (1991) Ann. Statist, 19, pp. 229-248. , MR1091847
- MARONNA, R.A., Robust M-estimators of multivariate location and scatter (1976) Ann. Statist, 4, pp. 51-67. , MR0388656
- MARONNA, R. A. andYOHAI, V. J. (2008). Robust lower-rank approximation of data matrices with element-wise contamination. Technometrics 50 295-304; MARTIN, R.D., YOHAI, V.J., ZAMAR, R.H., Min-max bias robust regression (1989) Ann. Statist, 17, pp. 1608-1630. , MR1026302
- ROUSSEEUW, P.J., Least median of squares regression (1984) J. Amer. Statist. Assoc, 79, pp. 871-880. , MR0770281
- STAHEL, W.A., (1981) Robuste Schätzungen: Infinitesimale Optimalität und Schätzungen von Kovarianzmatrizen, , PhD thesis, ETH Zürich
- TATSUOKA, K.S., TYLER, D.E., The uniqueness of S and M-functionals under nonelliptical distributions (2000) Ann. Statist, 28, pp. 1219-1243. , MR1811326
- TUKEY, J.W., The future of data analysis (1962) Ann. Math. Statist, 33, pp. 1-67. , MR0133937
- TYLER, D.E., High-breakdown point multivariate M-estimation (2002) Estadística, 54, pp. 213-247. , MR2022808
Citas:
---------- APA ----------
Alqallaf, F., Van Aelst, S., Yohai, V.J. & Zamar, R.H.
(2009)
. Propagation of outliers in multivariate data. Annals of Statistics, 37(1), 311-331.
http://dx.doi.org/10.1214/07-AOS588---------- CHICAGO ----------
Alqallaf, F., Van Aelst, S., Yohai, V.J., Zamar, R.H.
"Propagation of outliers in multivariate data"
. Annals of Statistics 37, no. 1
(2009) : 311-331.
http://dx.doi.org/10.1214/07-AOS588---------- MLA ----------
Alqallaf, F., Van Aelst, S., Yohai, V.J., Zamar, R.H.
"Propagation of outliers in multivariate data"
. Annals of Statistics, vol. 37, no. 1, 2009, pp. 311-331.
http://dx.doi.org/10.1214/07-AOS588---------- VANCOUVER ----------
Alqallaf, F., Van Aelst, S., Yohai, V.J., Zamar, R.H. Propagation of outliers in multivariate data. Ann. Stat. 2009;37(1):311-331.
http://dx.doi.org/10.1214/07-AOS588