Robust and efficient estimation of multivariate scatter and location

Several equivariant estimators of multivariate location and scatter are studied, which are highly robust, have a controllable finite-sample efficiency and are computationally feasible in large dimensions. The most frequently employed estimators are not quite satisfactory in this respect. The Minimum...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: Maronna, R.A
Otros Autores: Yohai, V.J
Formato: Capítulo de libro
Lenguaje:Inglés
Publicado: Elsevier B.V. 2017
Acceso en línea:Registro en Scopus
DOI
Handle
Registro en la Biblioteca Digital
Aporte de:Registro referencial: Solicitar el recurso aquí
LEADER 08686caa a22008057a 4500
001 PAPER-15019
003 AR-BaUEN
005 20230518204538.0
008 190410s2017 xx ||||fo|||| 00| 0 eng|d
024 7 |2 scopus  |a 2-s2.0-85007109965 
040 |a Scopus  |b spa  |c AR-BaUEN  |d AR-BaUEN 
030 |a CSDAD 
100 1 |a Maronna, R.A. 
245 1 0 |a Robust and efficient estimation of multivariate scatter and location 
260 |b Elsevier B.V.  |c 2017 
270 1 0 |m Maronna, R.A.; Departamento de Matemática, Facultad de Ciencias Exactas, C.C. 172, Argentina; email: rmaronna@retina.ar 
506 |2 openaire  |e Política editorial 
504 |a Agostinelli, C., Leung, A., Yohai, V.J., Zamar, R.H., Robust estimation of multivariate location and scatter in the presence of cellwise and casewise contamination (2015) TEST, 24, pp. 441-461 
504 |a Croux, C., Haesbroeck, G., Influence function and efficiency of the minimum covariance determinant scatter matrix estimator (1999) J. Multivariate Anal., 71, pp. 161-190 
504 |a Davies, P.L., Asymptotic behaviour of S-estimates of multivariate location parameters and dispersion matrices (1987) Ann. Statist., 15, pp. 1269-1292 
504 |a Donoho, D., Breakdown properties of multivariate location estimators (1982), (Ph.D. thesis) Harvard University unpublished thesis; Gnanadesikan, R., Kettenring, J.R., Robust estimates, residuals, and outlier detection with multiresponse data (1972) Biometrics, 28, pp. 81-124 
504 |a Hampel, F.R., Ronchetti, E.M., Rousseeuw, P.J., Stahel, W.A., Robust Statistics: The Approach Based on Influence Functions (1986), John Wiley & Sons; Hubert, M., Rousseeuw, P., Vanpaemel, D., Verdonck, T., The DetS and DetMM estimators for multivariate location and scatter (2015) Data Anal., 81, pp. 64-75 
504 |a Janssens, K.H., Deraedt, I., Schalm, O., Veeckman, J., Composition of 15–17th Century Archaeological Glass Vessels Excavated in Antwerp, Belgium (1998), pp. 253-267. , Springer Vienna Vienna; Kent, J.T., Tyler, D.E., Constrained M-estimation for multivariate location and scatter (1996) Ann. Statist., 24, pp. 1346-1370 
504 |a Locantore, N., Marron, J., Simpson, D., Tripoli, N., Zhang, J., Cohen, K., Robust principal component analysis for functional data (1999) TEST, 8, pp. 1-73 
504 |a Lopuhaä, H.P., Multivariate τ-estimators for location and scatter (1991) Canad. J. Statist., 19, pp. 307-321 
504 |a Lopuhaä, H.P., Highly efficient estimators of multivariate location with high breakdown point (1992) Ann. Statist., 20, pp. 398-413 
504 |a Maronna, R.A., Robust M-estimators of multivariate location and scatter (1976) Ann. Statist., 4, pp. 51-67 
504 |a Maronna, R.A., Martin, D.R., Yohai, V.J., Robust Statistics: Theory and Methods (2006), Wiley; Maronna, R.A., Yohai, V.J., The behavior of the Stahel–Donoho robust multivariate estimator (1995) J. Amer. Statist. Assoc., 90, pp. 330-341 
504 |a Muler, N., Yohai, V., Robust estimates for ARCH processes (2002) J. Time Series Anal., 23, pp. 341-375 
504 |a Paindaveine, D., Bever, G.V., Inference on the shape of elliptical distributions based on the MCD (2014) J. Multivariate Anal., 129, pp. 125-144 
504 |a Peña, D., Prieto, F., Combining random and specific directions for outlier detection and robust estimation in high-dimensional multivariate data (2007) J. Comput. Graph. Statist., 16, pp. 228-254 
504 |a Rocke, D.M., Robustness properties of S-estimators of multivariate location and shape in high dimension (1996) Ann. Statist., 24, pp. 1327-1345 
504 |a Rousseeuw, P., Multivariate Estimation with High Breakdown Point (1985), pp. 283-297. , Reidel Publishing Company Dordrecht; Stahel, W., Breakdown of covariance estimators. Tech. Rep. (1981), E.T.H. Zürich; Tatsuoka, K.S., Tyler, D.E., On the uniqueness of S-functionals and M-functionals under nonelliptical distributions (2000) Ann. Statist., 28, pp. 1219-1243 
504 |a Tyler, D.E., Finite sample breakdown points of projection based multivariate location and scatter statistics (1994) Ann. Statist., 22, pp. 1024-1044 
504 |a Verboven, S., Hubert, M., Libra: a MATLAB library for robust analysis (2005) Chemometr. Intell. Lab. Syst., 75, pp. 127-136 
504 |a Yohai, V.J., High breakdown-point and high efficiency robust estimates for regression (1987) Ann. Statist., 15, pp. 642-656 
504 |a Yohai, V.J., Zamar, R.H., High breakdown-point estimates of regression by means of the minimization of an efficient scale (1988) J. Amer. Statist. Assoc., 83, pp. 406-413 
504 |a Zuo, Y., Cui, H., He, X., On the Stahel–Donoho estimator and depth-weighted means of multivariate data (2004) Ann. Statist., 32, pp. 167-188 
520 3 |a Several equivariant estimators of multivariate location and scatter are studied, which are highly robust, have a controllable finite-sample efficiency and are computationally feasible in large dimensions. The most frequently employed estimators are not quite satisfactory in this respect. The Minimum Volume Ellipsoid (MVE) and the Minimum Covariance Determinant (MCD) estimators are known to have a very low efficiency. S-estimators with a monotonic weight function like the bisquare have a low efficiency when the dimension p is small, and their efficiency tends to one with increasing p. Unfortunately, this advantage is outweighed by a serious loss in robustness for large p. Four families of estimators with controllable efficiencies whose performance for moderate to large p has not been explored to date are studied: S-estimators with a non-monotonic weight function, MM-estimators, τ-estimators, and the Stahel–Donoho estimator. Two types of starting estimators are employed: the MVE computed through subsampling, and a semi-deterministic procedure previously proposed for outlier detection, based on the projections with maximum and minimum kurtosis. A simulation study shows that an S-estimator with non-monotonic weight function can simultaneously attain high efficiency and high robustness for p≥15, while an MM-estimator with a particular weight function can be recommended for p>15. For both recommended estimators, the initial values are given by the semi-deterministic procedure mentioned above. © 2016 Elsevier B.V.  |l eng 
536 |a Detalles de la financiación: Universidad de Buenos Aires 
536 |a Detalles de la financiación: Agencia Nacional de Promoción Científica y Tecnológica, PIP 112-2011-01- 00339 
536 |a Detalles de la financiación: Consejo Nacional de Investigaciones Científicas y Técnicas, 20020130100279BA 
536 |a Detalles de la financiación: This work has been partially supported by grants PICT 2011-0397 from ANPCyT, PIP 112-2011-01- 00339 from CONICET and 20020130100279BA from Universidad de Buenos Aires. The authors thank Professors Daniel Pe?a and Javier Prieto for providing them with the MATLAB code to compute the KSD estimator. 
593 |a Department of Mathematics, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, Argentina 
593 |a Department of Mathematics, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires and CONICET, Argentina 
690 1 0 |a KULLBACK–LEIBLER DIVERGENCE 
690 1 0 |a MM-ESTIMATOR 
690 1 0 |a S-ESTIMATOR 
690 1 0 |a STAHEL–DONOHO ESTIMATOR 
690 1 0 |a Τ-ESTIMATOR 
690 1 0 |a MULTIVARIABLE SYSTEMS 
690 1 0 |a SAMPLING 
690 1 0 |a EFFICIENT ESTIMATION 
690 1 0 |a EQUIVARIANT ESTIMATORS 
690 1 0 |a MINIMUM COVARIANCE DETERMINANT 
690 1 0 |a MINIMUM VOLUME ELLIPSOIDS 
690 1 0 |a MM-ESTIMATOR 
690 1 0 |a OUTLIER DETECTION 
690 1 0 |a S-ESTIMATORS 
690 1 0 |a SIMULATION STUDIES 
690 1 0 |a EFFICIENCY 
700 1 |a Yohai, V.J. 
773 0 |d Elsevier B.V., 2017  |g v. 109  |h pp. 64-75  |p Comput. Stat. Data Anal.  |x 01679473  |w (AR-BaUEN)CENRE-4276  |t Computational Statistics and Data Analysis 
856 4 1 |u https://www.scopus.com/inward/record.uri?eid=2-s2.0-85007109965&doi=10.1016%2fj.csda.2016.11.006&partnerID=40&md5=a53366b81e151fbee141ab7f34019ebf  |y Registro en Scopus 
856 4 0 |u https://doi.org/10.1016/j.csda.2016.11.006  |y DOI 
856 4 0 |u https://hdl.handle.net/20.500.12110/paper_01679473_v109_n_p64_Maronna  |y Handle 
856 4 0 |u https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_01679473_v109_n_p64_Maronna  |y Registro en la Biblioteca Digital 
961 |a paper_01679473_v109_n_p64_Maronna  |b paper  |c PE 
962 |a info:eu-repo/semantics/article  |a info:ar-repo/semantics/artículo  |b info:eu-repo/semantics/publishedVersion 
999 |c 75972