Advertisement

Time Series Predictions Using Multi-scale Support Vector Regressions

  • Danian Zheng
  • Jiaxin Wang
  • Yannan Zhao
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3959)

Abstract

Support vector regressions (SVR) have been applied to time series prediction recently and perform better than RBF networks. However, only one kernel scale is used in SVR. We implemented a multi scale support vector regression (MS-SVR), which has several different kernel scales, and tested it on two time series benchmarks: Mackey-Glass time series and Laser generated data. In both cases, MS-SVR improves the performance of SVR greatly: fewer support vectors and less prediction error.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Vapnik, V.N., Golowich, S.E., Smola, A.: Support Vector Method for Function Approximation, Regression Estimation and Signal Processing. Advances in Neural Information Processing System 9, 281–287 (1997)Google Scholar
  2. Müller, K.R., Smola, A.J., Rätsch, G., Schölkopf, B., Kohlmorgen, J., Vapnik, V.: Predicting Time Series with Support Vector Machines. In: Proc. of the 7th International Conference on Artificial Neural Networks, pp. 999–1004 (1997)Google Scholar
  3. Mukherjee, S., Osuna, E., Girosi, F.: Nonlinear Prediction of Chaotic Time Series Using Support Vector Machines. In: Proc. of the IEEE Signal Processing Society Workshop in Neural Networks for Signal Processing VII, pp. 511–520 (1997)Google Scholar
  4. Tronci, S., Giona, M., Baratti, R.: Reconstruction of chaotic time series by neural models: a case study. Neurocomputing 55, 581–591 (2003)CrossRefGoogle Scholar
  5. Zheng, D.N., Wang, J.X., Zhao, Y.N.: Non-flat function estimation with a multiscale support vector regression (2005) (submitted)Google Scholar
  6. Cristianini, N., Sharwe-Tayor, J.: An Introduction to Support Vector Machines. Cambridge University Press, Cambridge (2000)Google Scholar
  7. Figueiredo, M.A.T.: Adaptive Sparseness Using Jeffreys Prior. In: Advances in Neural Information Processing System, vol. 14, pp. 697–704 (2001)Google Scholar
  8. Suykens, J.A.K., Vandewalle, J.: Least Squares Support Vector Machine Classifiers. Neural Processing Letters 9(3), 293–300 (1999)CrossRefMathSciNetGoogle Scholar
  9. Cao, L.J.: Support vector machines experts for time series forecasting. Neurocomputing 51, 321–339 (2003)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Danian Zheng
    • 1
  • Jiaxin Wang
    • 1
  • Yannan Zhao
    • 1
  1. 1.Department of Computer Science and TechnologyTsinghua UniversityBeijingChina

Personalised recommendations