In this paper, we consider a robust version of multiple-set linear canonical analysis obtained by using a S-estimator of covariance operator. The related influence functions are derived. Asymptotic properties of this robust method are obtained and a robust test for mutual non-correlation is introduced.
Dans cet article, nous considérons une version robuste de l’analyse canonique linéaire généralisée obtenue en utilisant un S-estimateur de l’opérateur de covariance. Les fonctions d’influence correspondantes sont déterminées. Les propriétés asymptotiques de cette méthode robuste sont obtenues, et un test robuste de non-corrélation mutuelle est introduit.
Revised:
Accepted:
Published online:
Ulrich Djemby Bivigou 1; Guy Martial Nkiet 1
@article{CRMATH_2020__358_5_571_0, author = {Ulrich Djemby Bivigou and Guy Martial Nkiet}, title = {Robustifying multiple-set linear canonical analysis with {S-estimator}}, journal = {Comptes Rendus. Math\'ematique}, pages = {571--576}, publisher = {Acad\'emie des sciences, Paris}, volume = {358}, number = {5}, year = {2020}, doi = {10.5802/crmath.74}, language = {en}, }
TY - JOUR AU - Ulrich Djemby Bivigou AU - Guy Martial Nkiet TI - Robustifying multiple-set linear canonical analysis with S-estimator JO - Comptes Rendus. Mathématique PY - 2020 SP - 571 EP - 576 VL - 358 IS - 5 PB - Académie des sciences, Paris DO - 10.5802/crmath.74 LA - en ID - CRMATH_2020__358_5_571_0 ER -
Ulrich Djemby Bivigou; Guy Martial Nkiet. Robustifying multiple-set linear canonical analysis with S-estimator. Comptes Rendus. Mathématique, Volume 358 (2020) no. 5, pp. 571-576. doi : 10.5802/crmath.74. https://comptes-rendus.academie-sciences.fr/mathematique/articles/10.5802/crmath.74/
[1] nAalyse canonique basée sur des estimateurs robustes de la matrice des covariances, Rev. Stat. Appl., Volume 50 (2002), pp. 5-26
[2] Classification efficiencies for robust linear discriminant analysis, Stat. Sin., Volume 18 (2008) no. 2, pp. 581-599 | MR | Zbl
[3] Principal component analysis based on robust estimators of the covariance or correlation matrix: Influence function and efficiencies, Biometrika, Volume 87 (2000) no. 3, pp. 603-618 | DOI | MR | Zbl
[4] Asymptotic behaviour of S-estimates of multivariate location and parameters and dispersion matrices, Ann. Stat., Volume 15 (1987), pp. 1269-1292 | DOI | MR | Zbl
[5] A synthesis of canonical variate analysis, generalised canonical correlation and Procrustes analysis, Comput. Stat. Data Anal., Volume 50 (2006) no. 1, pp. 107-134 | DOI | MR | Zbl
[6] Nonlinear multivariate analysis, Wiley Series in Probability and Mathematical Statistics, John Wiley & Sons, 1990 | MR | Zbl
[7] Robust Statistics: The Approach based on influence functions, Wiley Series in Probability and Mathematical Statistics, John Wiley & Sons, 1986 | Zbl
[8] Functional multiple-set canonical correlation analysis, Psychometrika, Volume 77 (2012), pp. 753-775 | MR | Zbl
[9] On the relation between S-estimator and M-estimator of multivariate location and covariance, Ann. Stat., Volume 17 (1989) no. 4, pp. 1662-1683 | DOI | MR | Zbl
[10] Asymptotic expansion of S-estimators of location and covariance, Stat. Neerl., Volume 51 (1997) no. 2, pp. 220-237 | DOI | MR | Zbl
[11] Asymptotic theory of multiple-set linear canonical analysis, Math. Methods Stat., Volume 26 (2017) no. 3, pp. 196-211 | DOI | MR | Zbl
[12] Regularized multiple-set canonical correlation analysis, Psychometrika, Volume 73 (2008) no. 4, pp. 753-775 | DOI | MR | Zbl
[13] Robustifying principal component analysis with spatial sign vectors, Stat. Probab. Lett., Volume 82 (2012) no. 4, pp. 765-774 | DOI | MR | Zbl
[14] Regularized generalized canonical correlation analysis, Psychometrika, Volume 76 (2011) no. 2, pp. 257-284 | DOI | MR | Zbl
Cited by Sources:
Comments - Policy