Abstract
A class of robust methods for multivariate data analysis was proposed to enable processing of data that do not satisfy assumptions necessary for the application of classical data analysis methods. Canonical covariance analysis (Momirović, Dobrić and Karaman, 1983) was developed as the general method to analyse relationships between two sets of variables. A model of robust redundancy analysis (Prot, Bosnar and Momirović, 1983), as well as models of robust regression (Štalec and Momirović, 1983) and discriminant (Dobrić and Momirović, 1984) analysis, were derived as special cases from the canonical covariance model. In order to clarify the interpretation of the obtained results, relationships between robust methods and corresponding classical methods were described in our previous papers. A set of programs for all proposed methods was written in GENSTAT and SS languages, and the behaviour of the methods was examined on real data.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Dobrić V. and K. Momirović (1984), An algorithm and program for robust discriminant analysis, Proceedings of 8th Bosnian symposium in informatics “Jahorina 84”, Sarajevo, 213 1–5.
Dobrić V., J. Štalec and K. Momirović (1984), Note on some relationships between least squares and robust regression analysis, Proceedings of 6th international symposium “Computer at the University”, Dubrovnik, 507 1–7.
Dobrić V., K. Momirović and M. Gredelj (1985), Quasicanonical relationships of variables in universal metric space, Proceedings of 7th International symposium “Computer at the University”, Cavtat, 518 1–6.
Gredelj M., K. Momirović and V. Dobrić (1986), Some relations between canonical covariance analysis and principal component analysis, Proceeding of 8th International symposium “Computer at the University”, Cavtat, 5.03 1–5.
Lužar V. (1985), Inference in quasicanonical correlation analysis, Proceeding of 7th International symposium “Computer at the University”, Cavtat, 518 1–6.
Lužar V. (1986), Inference in robust discriminant analysis, Proceeding of 8th International symposium “Computer at the University”, Cavtat, 5.08 1–9.
Momirović K., V. Dobrić and Š. Karaman (1983), Canonical covariance analysis, Proceedings of 5th International symposium “Computer at the University”, Cavtat, 463–473.
Momirović K., V. Dobrić and M. Gredelj (1985), Quasicanonical relationships of partial image variables, Proceedings of 7th International symposium “Computer at the University”, Cavtat, 522 1–6.
Momirović K. and V. Dobrić (1985), Diagnostic efficiency of a robust method for discriminant analysis. Abstracts of 1st European Biometric Conference, Hungarian Academy of sciences, Martonvasar, p. 101.
Momirović K. and V. Dobrić (1985), Some relations between canonical and quasicanonical correlation analysis, Proceedings of 4th international symposium “Data analysis and Informatics”, Paris, Posters, 101–104.
Prot F., K. Bosnar and K. Momirović (1983), An algorithm for assymetric redundancy analysis without orthogonality constraints, Proceedings of 5th International symposium “Computer at the University”, Cavtat, 485–490.
Štalec J. and K. Momirović (1983), Some properties of a very simple model for robust regression analysis, Proceedings of 5th international symposium “Computer at the University”, Cavtat, 453–461.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1986 Physica-Verlag, Heidelberg for IASC (International Association for Statistical Computing)
About this paper
Cite this paper
Dobrić, V. (1986). On a Class of Robust Methods for Multivariate Data Analysis. In: De Antoni, F., Lauro, N., Rizzi, A. (eds) COMPSTAT. Physica-Verlag HD. https://doi.org/10.1007/978-3-642-46890-2_31
Download citation
DOI: https://doi.org/10.1007/978-3-642-46890-2_31
Publisher Name: Physica-Verlag HD
Print ISBN: 978-3-7908-0355-6
Online ISBN: 978-3-642-46890-2
eBook Packages: Springer Book Archive