A State Space Approach and Hurst Exponent for Ensemble Predictors

  • Ryszard Szupiluk
  • Tomasz Ząbkowski
Part of the Communications in Computer and Information Science book series (CCIS, volume 383)


In this article we propose a concept of ensemble methods based on deconvolution with state space and MLP neural network approach. Having a few prediction models we treat their results as a multivariate variable with latent components having destructive or constructive impact on prediction. The latent component classification is performed using novel variability measure derived from Hurst exponent. The validity of our concept is presented on the real problem of load forecasting in the Polish power system.


state space approach Hurst exponent Independent Component Analysis ensemble methods 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Barabasi, A.-L., Vicsek, T.: Multifractality of self-affine fractals. Physical Review A44, 2730–2733 (1991)Google Scholar
  2. 2.
    Breiman, L.: Bagging predictors. Machine Learning 24, 123–140 (1996)MathSciNetzbMATHGoogle Scholar
  3. 3.
    Bussgang, J.J.: Cross-correlation function of amplitude-distorted Gaussian signals, MIT Research Laboratory of Electronics, Technical Report 216 (1952)Google Scholar
  4. 4.
    Cichocki, A., Amari, S.: Adaptive Blind Signal and Image Processing. John Wiley, Chichester (2002)CrossRefGoogle Scholar
  5. 5.
    Hoeting, J., Mdigan, D., Raftery, A., Volinsky, C.: Bayesian model averaging: a tutorial. Statistical Science 14, 382–417 (1999)MathSciNetCrossRefzbMATHGoogle Scholar
  6. 6.
    Hyvärinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. John Wiley, Chichester (2001)CrossRefGoogle Scholar
  7. 7.
    Hurst, H.E.: Long term storage capacity of reservoirs. Trans. Am. Soc. Civil Engineers 116, 770–799 (1951)Google Scholar
  8. 8.
    Peters, E.: Fractal market analysis. John Wiley, Chichester (1996)Google Scholar
  9. 9.
    Samorodnitskij, G., Taqqu, M.: Stable non-Gaussian random processes: stochastic models with infinitive variance. Chapman and Hall, New York (1994)Google Scholar
  10. 10.
    Szupiluk, R., Wojewnik, P., Ząbkowski, T.: Smooth Component Analysis as Ensemble Method for Prediction Improvement. In: Davies, M.E., James, C.J., Abdallah, S.A., Plumbley, M.D. (eds.) ICA 2007. LNCS, vol. 4666, pp. 277–284. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  11. 11.
    Szupiluk, R., Wojewnik, P., Zabkowski, T.: Noise detection for ensemble methods. In: Rutkowski, L., Scherer, R., Tadeusiewicz, R., Zadeh, L.A., Zurada, J.M. (eds.) ICAISC 2010, Part I. LNCS (LNAI), vol. 6113, pp. 471–478. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  12. 12.
    Weron, R., Bierbrauer, M., Trück, S.: Modeling electricity prices: jump diffusion and regime switching. Physica A 336, 39 (2004)CrossRefGoogle Scholar
  13. 13.
    Zhang, L., Cichocki, A.: Blind Separation of Filtered Source Using State-Space Approach. In: Advances in Neural Information Processing Systems, vol. 11, pp. 648–654 (1999)Google Scholar
  14. 14.
    Zhang, L., Cichocki, A.: Blind Deconvolution of Dynamical Systems: A State Space Approach. Journal of Signal Processing 4(2), 111–113 (2000)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Ryszard Szupiluk
    • 1
  • Tomasz Ząbkowski
    • 2
  1. 1.Warsaw School of EconomicsWarsawPoland
  2. 2.Warsaw University of Life SciencesWarsawPoland

Personalised recommendations