Abstract:
All known approaches to nonlinear principal components are based on minimizing a quadratic loss, which makes them sensitive to data contamination. A predictive approach in which a spline curve is fit minimizing a residual M-scale is proposed for this problem. For a p-dimensional random sample xi (i=1,…,n) the method finds a function h:R→Rp and a set {t1,…,tn}⊂R that minimize a joint M-scale of the residuals xi−h(ti), where h ranges on the family of splines with a given number of knots. The computation of the curve then becomes the iterative computing of regression S-estimators. The starting values are obtained from a robust linear principal components estimator. A simulation study and the analysis of a real data set indicate that the proposed approach is almost as good as other proposals for row-wise contamination, and is better for element-wise contamination. © 2013, Springer Science+Business Media New York.
Registro:
Documento: |
Artículo
|
Título: | Robust nonlinear principal components |
Autor: | Maronna, R.A.; Méndez, F.; Yohai, V.J. |
Filiación: | University of La Plata, C.C. 172, La Plata, 1900, Argentina University of Rosario, Bv. Oroño 1261, Rosario, 2000, Argentina Departamento de Matemática, Universidad de Buenos Aires, Ciudad Universitaria, Pabellon 1, Buenos Aires, 1428, Argentina
|
Palabras clave: | Principal curves; S-estimators; Splines |
Año: | 2013
|
Volumen: | 25
|
Número: | 2
|
Página de inicio: | 439
|
Página de fin: | 448
|
DOI: |
http://dx.doi.org/10.1007/s11222-013-9442-0 |
Título revista: | Statistics and Computing
|
Título revista abreviado: | Stat. Comput.
|
ISSN: | 09603174
|
Registro: | https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_09603174_v25_n2_p439_Maronna |
Referencias:
- Alqallaf, F., Van Aelst, S., Yohai, V.J., Zamar, R.H., Propagation of outliers in multivariate data (2009) Ann. Stat, 37, pp. 311-331
- Bolton, R.J., Hand, D.J., Webb, A.R., Projection techniques for nonlinear principal components analysis (2003) Stat. Comput, 13, pp. 267-276
- Candès, E., Li, X., Ma, Y., Wright, J., Robust principal component analysis (2011) J. ACM, 58 (3)
- Cleveland, W., Robust locally weighted regression and smoothing scatterplots (1979) J. Am. Stat. Assoc, 74, pp. 829-836
- Croux, C., Filzmoser, P., Pison, G., Rousseeuw, P.J., Fitting multiplicative models by robust alternating regressions (2003) Stat. Comput, 13, pp. 23-36
- Delicado, P., Another look at principal curves and surfaces (2001) J. Multivar. Anal, 77, pp. 84-116
- Ein-Dor, P., Feldmesser, J., Attributes of the performance of central processing units: a relative performance prediction model (1987) Commun. ACM, 30, pp. 308-317
- Gerber, S., Whitaker, R., Regularization free principal curve estimation (2013) J. Mach. Learn. Res, 14, pp. 1285-1302
- Hastie, T., Stuetzle, W., Principal curves (1989) J. Am. Stat. Assoc, 84, pp. 502-516
- Hubert, M., Rousseeuw, P.J., Verboven, S., Robust PCA for high-dimensional data (2003) Developments in Robust Statistics, pp. 169-179. , Dutter R., Filzmoser P., Gather U., Rousseeuw P.J., (eds), Physika Verlag, Heidelberg:
- Locantore, N., Marron, J.S., Simpson, D.G., Tripoli, N., Zhang, J.T., Cohen, K.L., Robust principal components for functional data (1999) Test, 8, pp. 1-28
- Maronna, R., Principal components and orthogonal regression based on robust scales (2005) Technometrics, 47, pp. 264-273
- Maronna, R.A., Martin, R.D., Yohai, V.J., (2006) Robust Statistics: Theory and Methods, , Wiley, New York:
- Maronna, R.A., Yohai, V.J., Robust lower-rank approximation of data matrices with element-wise contamination (2008) Technometrics, 50, pp. 295-304
- Rousseeuw, P.J., Yohai, V.J., Robust regression by means of S estimators (1984) Robust and Nonlinear Time Series Analysis, pp. 256-272. , Franke J., Härdle W., Martin D., (eds), Lecture Notes in Statistics, 26, Springer, New York:
- Tharmaratnam, K., Claeskens, G., Croux, C., Salibian-Barrera, M., S-estimation for penalized regression splines (2010) J. Comput. Graph. Stat, 19, pp. 609-625
- Tibshirani, R., Principal curves revisited (1992) Stat. Comput, 2, pp. 183-190
- Verbeek, J.J., Vlassis, N., Kröse, B., A k-segments algorithm for finding principal curves (2002) Pattern Recognit. Lett, 23, pp. 1009-1017
- Yohai, V.J., High breakdown-point and high efficiency estimates for regression (1987) Ann. Stat, 15, pp. 642-665
- Yohai, V.J., Ackerman, W., Haigh, C., Nonlinear principal components (1985) Qual. Quant, 19, pp. 53-71
- Yohai, V.J., Zamar, R., High breakdown point estimates of regression by means of the minimization of an efficient scale (1988) J. Am. Stat. Assoc, 86, pp. 403-413
Citas:
---------- APA ----------
Maronna, R.A., Méndez, F. & Yohai, V.J.
(2013)
. Robust nonlinear principal components. Statistics and Computing, 25(2), 439-448.
http://dx.doi.org/10.1007/s11222-013-9442-0---------- CHICAGO ----------
Maronna, R.A., Méndez, F., Yohai, V.J.
"Robust nonlinear principal components"
. Statistics and Computing 25, no. 2
(2013) : 439-448.
http://dx.doi.org/10.1007/s11222-013-9442-0---------- MLA ----------
Maronna, R.A., Méndez, F., Yohai, V.J.
"Robust nonlinear principal components"
. Statistics and Computing, vol. 25, no. 2, 2013, pp. 439-448.
http://dx.doi.org/10.1007/s11222-013-9442-0---------- VANCOUVER ----------
Maronna, R.A., Méndez, F., Yohai, V.J. Robust nonlinear principal components. Stat. Comput. 2013;25(2):439-448.
http://dx.doi.org/10.1007/s11222-013-9442-0