Robust canonical correlation analysis: a predictive approach.

We present a method for robust Canonical Correlation Analysis based on a prediction approach. The robust canonical coordinates are obtained using robust estimators for the multivariate linear model. Two different methods are proposed to robustly estimate the canonical correlations. The performance of the proposed method is compared with those of the classical approach and of other robust estimators of Canonical Correlations Analysis through real and simulated data sets.

[1]  Pierre Duchesne,et al.  Robust estimation of the SUR model , 2000 .

[2]  P. Rousseeuw.  Multivariate estimation with high breakdown point , 1985 .

[3]  Z. Bai,et al.  ON CONSTRAINED M-ESTIMATION AND ITS RECURSIVE ANALOG IN MULTIVARIATE LINEAR REGRESSION MODELS , 2008 .

[4]  Yuehua Wu,et al.  A note on constrained M-estimation and its recursive analog in multivariate linear regression models , 2009 .

[5]  K. Janssens,et al.  Composition of 15-17th century archaeological glass vessels excavated in Antwerp, Belgium , 1998 .

[6]  J. Friedman,et al.  Predicting Multivariate Responses in Multiple Linear Regression , 1997 .

[7]  M. Victoria-Feser,et al.  A Robust Coefficient of Determination for Regression , 2010 .

[8]  Hendrik P. Lopuhaä,et al.  Highly efficient estimators of multivariate location with high breakdown point , 1992 .

[9]  Hannu Oja,et al.  Influence functions and efficiencies of the canonical correlation and vector estimates based on scatter and shape matrices , 2006 .

[10]  P. Sen,et al.  Restricted canonical correlations , 1994 .

[11]  Victor J. Yohai,et al.  Canonical Variables as Optimal Predictors , 1980 .

[12]  A. Izenman.  Reduced-rank regression for the multivariate linear model , 1975 .

[13]  Catherine Dehon,et al.  Outlier resistant estimators for canonical correlation analysis , 2000 .

[14]  Ricardo A. Maronna,et al.  Estimates of MM type for the multivariate linear model , 2010, J. Multivar. Anal..

[15]  David E. Tyler,et al.  Constrained M-estimation for multivariate location and scatter , 1996 .

[16]  Frederick R. Forst,et al.  On robust estimation of the location parameter , 1980 .

[17]  David E. Tyler,et al.  On the uniqueness of S-functionals and M-functionals under nonelliptical distributions , 2000 .

[18]  D. Brillinger.  Time series - data analysis and theory , 1981, Classics in applied mathematics.

[19]  V. Yohai,et al.  Robust Statistics: Theory and Methods , 2006 .

[20]  Catherine Dehon,et al.  Analyse canonique basée sur des estimateurs robustes de la matrice de covariance , 2002 .

[21]  S. Van Aelst,et al.  Principal Components Analysis Based on Multivariate MM Estimators With Fast and Robust Bootstrap , 2006 .

[22]  H. Hotelling.  Relations Between Two Sets of Variates , 1936 .

[23]  H. P. Lopuhaä.  Multivariate τ‐estimators for location and scatter , 1991 .

[24]  David R. Brillinger,et al.  Time Series: Data Analysis and Theory. , 1982 .

[25]  Stefan Van Aelst,et al.  MULTIVARIATE REGRESSION S-ESTIMATORS FOR ROBUST ESTIMATION AND INFERENCE , 2005 .

[26]  Gerald Karnel.  Robust canonical correlation and correspondence analysis , 1991 .

[27]  M. Romanazzi.  Influence in canonical correlation analysis , 1992 .

[28]  Peter Filzmoser,et al.  Robust canonical correlations: A comparative study , 2005, Comput. Stat..

[29]  Catherine Dehon,et al.  Estimators of the multiple correlation coefficient: Local robustness and confidence intervals , 2003 .

[30]  V. Yohai.  HIGH BREAKDOWN-POINT AND HIGH EFFICIENCY ROBUST ESTIMATES FOR REGRESSION , 1987 .

[31]  G. Seber.  Multivariate observations / G.A.F. Seber , 1983 .

[32]  V. Yohai,et al.  Robust estimation for the multivariate linear model based on a τ-scale , 2006 .