A New Approach to Robust Partial Least Squares Regression Analysis

Esra Polat; Suleyman Gunay

doi:https://doi.org/10.14445/22315373/IJMTT-V9P524

International Journal of Mathematics Trends and Technology

Research Article | Open Access | Download PDF

Volume 9 | Number 3 | Year 2014 | Article Id. IJMTT-V9P524 | DOI : https://doi.org/10.14445/22315373/IJMTT-V9P524

A New Approach to Robust Partial Least Squares Regression Analysis

Esra Polat , Suleyman Gunay

Citation :

Esra Polat , Suleyman Gunay, "A New Approach to Robust Partial Least Squares Regression Analysis," International Journal of Mathematics Trends and Technology (IJMTT), vol. 9, no. 3, pp. 197-205, 2014. Crossref, https://doi.org/10.14445/22315373/IJMTT-V9P524

Abstract

Partial Least Squares Regression (PLSR) is a linear regression technique developed to relate many independent variables to one or several dependent variables. Robust methods are introduced to reduce or remove the effects of outlying data points. In the previous studies in robust PLSR field it has been mentioned that if the sample covariance matrix is properly robustified further robustification of the linear regression steps of the PLS1 algorithm (PLSR with univariate dependent variable) becomes unnecessary. Therefore, the purpose of this study is to propose a new approach to robust PLSR based on statistical procedures for covariance matrix robustification by selecting the well-known S-estimators. Both simulation results and an analysis on a real data set, which is used in robust PLSR literature frequently, showing the effectiveness, success in fitting to regular data points and predictive power of the new proposed robust PLSR method.

Keywords

robust partial least squares regression, robust covariance matrix, S-estimators, goodness-of-fit, prediction, efficiency.

References

[1] S. Engelen, M. Hubert, K. Vanden Branden, and S. Verboven, “Robust PCR and Robust PLSR: A comparative study”, In Theory and Applications of Recent Robust Methods (M. Hubert, G. Pison, A. Struyf and S. V. Aelst, eds.), Birkhäuser, Basel, pp. 105–117, 2004.
[2] J. A. Gil and R. Romera, “On robust partial least squares (PLS) methods”, Journal of Chemometrics, vol. 12, pp. 365-378, 1998.
[3] J. González, D. Peña, and R. Romera, “A robust partial least squares regression method with applications”, Journal of Chemometrics, vol. 23, pp. 78–90, 2009.
[4] A. J. Hardy, P. MacLaurin, S. J. Haswell, S. De Jong, and B. G. M. Vandeginste, “Double-case diagnostic for outliers identification”, Chemometrics and Intelligent Laboratory Systems , vol. 34, pp. 117-129, 1996.
[5] M. Hubert and K. Vanden Branden, “Robust methods for Partial Least Squares Regression”, Journal of Chemometrics, vol. 17, pp. 537-549, 2003.
[6] M. Hubert, P.J. Rousseeuw, and T. Verdonck, “A deterministic algorithm for robust location and scatter”, Journal of Computational and Graphical Statistics , vol. 21, pp. 618-637, 2012.
[7] T. Naes, “Multivariate calibration when the error covariance matrix is structured”, Technometrics, 27:3, pp. 301-311, 1985.
[8] E. Polat, “New Approaches in Robust Partial Least Squares Regression Analysis”, Ph.D Turkish Thesis, Hacettepe University Department of Statistics, Ankara, Turkey, March 2014.
[9] M. Riani, D. Perrotta, and F. Torti, “FSDA: A MATLAB toolbox for robust analysis and interactive data exploration”, Chemometrics and Intelligent Laboratory Systems , vol. 116, pp. 17–32, 2012.
[10] M. Salibian-Barrera, S. Van Aelst, and G. Willems, “PCA based on multivariate MM-estimators with fast and robust bootstrap”, Journal of the American Statistical Association , vol. 101, pp. 1198-1211, 2006.
[11] S. Serneels, C. Croux, P. Filzmoser, and P. J. Van Espen, “Partial Robust M-regression”, Chemometrics and Intelligent Laboratory Systems, vol. 79, pp. 55-64, 2005.