Modeling Dst with Recurrent EM Neural Networks

  • Derrick Takeshi Mirikitani
  • Lahcen Ouarbya
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5768)


Recurrent Neural Networks have been used extensively for space weather forecasts of geomagnetospheric disturbances. One of the major drawbacks for reliable forecasts have been the use of training algorithms that are unable to account for model uncertainty and noise in data. We propose a probabilistic training algorithm based on the Expectation Maximization framework for parameterization of the model which makes use of a forward filtering and backward smoothing Expectation step, and a Maximization step in which the model uncertainty and measurement noise estimates are computed. Through numerical experimentation it is shown that the proposed model allows for reliable forecasts and also outperforms other neural time series models trained with the Extended Kalman Filter, and gradient descent learning.


Solar Wind Root Mean Square Error Interplanetary Magnetic Field Geomagnetic Storm Extended Kalman Filter 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Axford, W.I., Hines, C.O.: A unifying theory of high-latitude geophysical phenomena and geomagnetic storms. Can. J. Phys. 39, 1433–1464 (1961)MathSciNetCrossRefGoogle Scholar
  2. 2.
    de Freitas, J.F.G., Niranjan, M., Gee, A.H.: Dynamic Learning with the EM Algorithm for Neural Networks. J. Vlsi. Signal. Proc. 26, 119–131 (2000)CrossRefzbMATHGoogle Scholar
  3. 3.
    Dungey, J.W.: Interplanetary magnetic field and the auroral zones. Phys. Rev. Lett. 26, 47–48 (2000)Google Scholar
  4. 4.
    Freeman, J., Natal, A., Reiff, P., Denig, W., Gussenhoven-Shea, S., Heinemann, M., Rich, F., Hairston, M.: The use of neural networks to predict magnetospheric parameters for input to a magnetospheric forecast model. In: Proceedings of Artificial Intelligence Applications in Solar-Terrestrial Physics Workshop, pp. 167–181 (1993)Google Scholar
  5. 5.
    Gonzales, W.D., Joselyn, J.A., Kamide, Y., Kroehl, H.W., Rostoker, G., Tsurutani, B.T., Vasyliunas, V.M.: What is a geomagnetic storm? J. Geophys. Res. 99, 5771–5792 (1994)CrossRefGoogle Scholar
  6. 6.
    Ghahramani, Z., Hinton, G.E.: Parameter estimation for linear dynamical systems. Technical Report, University of Toronto CRG-TR-96-2 (1996)Google Scholar
  7. 7.
    Lundstedt, H.: Neural Networks and prediction of solar-terrestrial effects. Planet. Space Sci. 40, 457–464 (1992)CrossRefGoogle Scholar
  8. 8.
    Lundstedt, H., Wintoft, P.: Prediction of geomagnetic storms from solar wind data with the use of a neural network. Ann Geophys. 12, 19–24 (1994)CrossRefGoogle Scholar
  9. 9.
    Lundstedt, H., Gleisner, H., Wintoft, P.: Operational forecasts of the geomagnetic Dst index. Geophys. Res. Lett. 29(24), 34–1–34–4 (2002)Google Scholar
  10. 10.
    Pallocchia, G., Amata, E., Consolini, G., Marcucci, M.F., Bertello, I.: Geomagnetic Dst index forecast based on IMF data only. Ann. Geophys. 24, 989–999 (2006)CrossRefGoogle Scholar
  11. 11.
    Puskorius, G.V., Feldkamp, L.A.: Neurocontrol of nonlinear dynamical systems with Kalman filter trained recurrent networks. IEEE T Neural Networks 5(2), 279–297 (1994)CrossRefGoogle Scholar
  12. 12.
    Rauch, H.E., Tung, F., Striebel, C.T.: Maximum Likelihood Estimates of Linear Dynamic Models. AIAA Journal 3(8), 1445–1450 (1965)CrossRefGoogle Scholar
  13. 13.
    Schottky, B., Saad, D.: Statistical mechanics of EKF learning in neural networks. J. Phys. A 32(9), 1605–1621 (1999)CrossRefzbMATHGoogle Scholar
  14. 14.
    Shumway, R.H., Stoffer, D.S.: An Approach To Time Series Smoothing and Forecasting Using the EM Algorithm. J. Time Ser. Anal. 4, 253–264 (1982)CrossRefzbMATHGoogle Scholar
  15. 15.
    Wei, H.L., Bilings, S.A., Balikhin, M.A.: Prediction of the Dst index using multiresolution wavelet models. J. Geophys. Res. 109(A7), A07212 (2004)Google Scholar
  16. 16.
    W.H.L., Bilings, S.A., Balikhin, M.A.: Forecasting the Geomagnetic Activity of teh Dst index using Multiscale Radial Basis Function Networks. In: Advances in Space Research, vol. 40, pp. 1863–1870 (2007)Google Scholar
  17. 17.
    Wu, J.G., Lundstedt, H.: Geomagnetic storm predictions from solar wind data with the use of dynamic neural networks. J. Geophys. Res. 102(A7), 14–255 (1997)Google Scholar
  18. 18.
    Williams, R.J., Zipser, D.: A learning algorithm for continuously running fully connected recurrent neural networks. Neural Computation 1, 270–280 (1989)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Derrick Takeshi Mirikitani
    • 1
  • Lahcen Ouarbya
    • 1
  1. 1.Department of Computer Science, Goldsmiths CollegeUniversity of London, New Cross, LondonLondonUK

Personalised recommendations