Neural Processing Letters

, Volume 22, Issue 3, pp 311–324 | Cite as

A Resampling Test for the Total Independence of Stationary Time Series: Application to the Performance Evaluation of ICA Algorithms

Article

Abstract

This paper addresses the independence testing of stationary time series. We develop a resampling test based on the Kankainen–Ushakov test of total independence. The resampling test, contrary to the original test, can be also applied to the data with a time-structure. The simulation studies demonstrate the good performance of the proposed test even with strongly autocorrelated time series. As an application, we consider biomedical signal processing and independent component analysis (ICA). The independence test can be used as a performance criterion for ICA algorithms. The practical example of performance evaluation deals with the ICA of electroencephalogram (EEG) data.

Keywords

EEG independence independent component analysis multivariate time series 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Amari, S. -I., Cichocki, A. and Yang, H. A new learning algorithm for blind signal separation In: Advances in Neural Information Processing Systems, Vol. 8. pp. 757–763, Cambridge MA, MIT Press, 1996.Google Scholar
  2. Bell, A.J., Sejnowski, T.J. 1995An information-maximization approach to blind separation and blind deconvolutionNeural Computation711291159Google Scholar
  3. Blum, J.R., Kiefer, J., Rosenblatt, M. 1961Distribution free tests of independence based on the sample distribution functionAnnals of Mathematical Statistics32485498MathSciNetGoogle Scholar
  4. Cardoso, J., Souloumiac, A. 1993Blind beamforming for non Gaussian signalsIEE- Proceedings-F140362370Google Scholar
  5. Chiu, K. -C., Liu, Z. -Y. and Xu, L. A statistical approach to testing mutual independence of ICA recovered sources. In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation (ICA2003), pp. 621–626, 2003.Google Scholar
  6. Comon, P. 1994Independent component analysis, a new conceptSignal Processing36287314CrossRefMATHGoogle Scholar
  7. Cotterill, D.S., Csörgő, M. 1985On the limiting distribution of and critical values for the Hoeffding, Blum, Kiefer, Rosenblatt independence criterionStatistics and Decision3148MathSciNetGoogle Scholar
  8. Csörgő, M. 1985Testing for independence by the empirical characteristic functionJournal of Multivariate Analysis16290299MathSciNetGoogle Scholar
  9. De Wet, T. 1980Cramer-von Mises tests for independenceJournal of Multivariate Analysis103850MATHMathSciNetGoogle Scholar
  10. Delorme, A., Makeig, S. et al. EEGLAB: Matlab Toolbox for Electrophysiological Research, www.sccn.ucsd.edu/eeglab, 2004.Google Scholar
  11. Eriksson, J., Koivunen, V. 2003aCharacteristic-function-based independent component analysisSignal Processing8321952208CrossRefGoogle Scholar
  12. Eriksson, J. and Koivunen, V. Identifiability and separability of linear ICA models revisited, In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation, pp. 23–27, 2003b.Google Scholar
  13. Fraser, A.M., Swinney, H.L. 1986Independent coordinates of strange attractors from mutual informationPhysical Review A3311341140CrossRefADSMathSciNetGoogle Scholar
  14. Giannakopoulos, X., Karhunen, J., Oja, E. 1999An experimental comparison of neural algorithms for independent component analysis and blind separationIntenational Journal of Neural Systems999114Google Scholar
  15. Gribonval, R., Benaroya, L., Vincent, E. and Fevotte, C. Proposals for performance measurement in source separation, In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation, pp. 715–720, 2003.Google Scholar
  16. Hegger, R., Kantz, H., Schreiber, T. 1999Practical implementation of nonlinear time series method: The TISEAN packageChaos9413435CrossRefADSMathSciNetGoogle Scholar
  17. Hoeffding, W. 1948A non-parametric test of independenceAnnals of Mathematical Statistics19546557MATHGoogle Scholar
  18. Hyvärinen, A. 1999Fast and robust fixed-point algorithms for independent component analysisIEEE Transactions on Neural Networks10626634Google Scholar
  19. Hyvärinen, A., Karhunen, J. and Oja, E. Independent Component Analysis, John Wiley & Sons, 2001.Google Scholar
  20. Ito, D., Mukai, T. and Murata, N. An approach of grouping decomposed components, In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation (ICA2003), pp 745–750, 2003.Google Scholar
  21. Jung, T.P., Makeig, S., Westerfield, M., Townsend, J., Courchesne, E., Sejnowski, T.J. 2001Analysis and visualization of single-trial event-related potentialsHuman Brain Mapping14166185CrossRefGoogle Scholar
  22. Kagan, A.M., Linnik, Y.V., Rao, C.R. 1973Characterization Problems in Mathematical StatisticsWileyNew YorkGoogle Scholar
  23. Kankainen, A. Consistent testing of total independence based on the empirical characteristic function, Ph.D. Thesis, University of Jyväskylä, 1995.Google Scholar
  24. Kankainen, A., Ushakov, N.G. 1998A consistent modification of a test for independence based on the empirical characteristic functionJournal of Mathematical Sciencies8915821589MathSciNetGoogle Scholar
  25. Karvanen, J., Koivunen, V. 2002‘Blind separation methods based on Pearson system its extensionsSignal Processing82663673CrossRefGoogle Scholar
  26. Makeig, S., Bell, A. J., Jung, T. -P. and Sejnowski, T. J. Independent component analysis of electroencephalographic data, In: Advances in Neural Information Processing Systems, Vol. 8, pp. 145–151, 1996.Google Scholar
  27. Meinecke, F., Ziehe, A., Kawanabe, M. and Müller, K. -R. Assessing reliability of ICA projections – a resampling approach, In: Proc. 3rd Int. Conference on Independent Component Analysis and Blind Signal Separation, ICA2001, pp. 74–79, 2001.Google Scholar
  28. Schobben, D., Torkkola, K. and Smaragdis, P. Evaluation of blind signal separation methods, In: Proc. Int. Workshop Independent Component Analysis and Blind Signal Separation, pp. 261–266, 1999.Google Scholar
  29. Ushakov, N. G. Selected Topics in Characteristic Functions, Modern Probability and Statistics VSP, 1999.Google Scholar
  30. Wilks, S.S. 1935The likelihood test of independence in contingency tablesAnnals of Mathematical Statistics6190196MATHGoogle Scholar

Copyright information

© Springer 2005

Authors and Affiliations

  1. 1.Laboratory for Advanced Brain Signal ProcessingRIKEN Brain Science InstituteJapan

Personalised recommendations