Kalman Filters and Neural Networks in Forecasting and Trading

  • Georgios Sermpinis
  • Christian Dunis
  • Jason Laws
  • Charalampos Stasinakis
Part of the Communications in Computer and Information Science book series (CCIS, volume 311)


The motivation of this paper is to investigate the use of a Neural Network (NN) architecture, the Psi Sigma Neural Network, when applied to the task of forecasting and trading the Euro/Dollar exchange rate and to explore the utility of Kalman Filters in combining NN forecasts. This is done by benchmarking the statistical and trading performance of PSN with a Naive Strategy and two different NN architectures, a Multi-Layer Perceptron and a Recurrent Network. We combine our NN forecasts with Kalman Filter, a traditional Simple Average and the Granger- Ramanathan’s Regression Approach. The statistical and trading performance of our models is estimated throughout the period of 2002-2010, using the last two years for out-of-sample testing. The PSN outperforms all models’ individual performances in terms of statistical accuracy and trading performance. The forecast combinations also present improved empirical evidence, with Kalman Filters outperforming by far its benchmarks.


Psi Sigma Network Recurrent Network Forecast Combinations Kalman Filter 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bates, J.M., Granger, C.W.J.: The Combination of Forecasts. Operational Research Society 20(4), 451–468 (1969)CrossRefGoogle Scholar
  2. 2.
    Deutsch, M., Granger, C.W.J., Teräsvirta, T.: The combination of forecasts using changing weights. International Journal of Forecasting 10(1), 47–57 (1994)CrossRefGoogle Scholar
  3. 3.
    Dunis, C.L., Huang, X.: Forecasting and trading currency volatility: an application of recurrent neural regression and model combination. Journal of Forecasting 21(5), 317–354 (2002)CrossRefGoogle Scholar
  4. 4.
    Dunis, C.L., Shannon, G.: Emerging Markets of South-East and Central Asia: Do they Still Offer a Diversification Benefit? Journal of Asset Management 6(3), 168–190 (2005)CrossRefGoogle Scholar
  5. 5.
    Dunis, C.L., Williams, M.: Modelling and Trading the EUR/USD Exchange Rate: Do Neural Network Models Perform Better? Derivatives Use, Trading and Regulation 8, 211–239 (2002)Google Scholar
  6. 6.
    Elman, J.L.: Finding Structure in Time. Cognitive Science 14(2), 179–211 (1990)CrossRefGoogle Scholar
  7. 7.
    Ghosh, J., Shin, Y.: The Pi-Sigma Network: An efficient Higher-order Neural Networks for Pattern Classification and Function Approximation. In: Proceedings of International Joint Conference of Neural Networks, vol. 1, pp. 13–18 (1991)Google Scholar
  8. 8.
    Ghosh, J., Shin, Y.: Efficient Higher-Order Neural Networks for Classification and Function Approximation. International Journal of Neural Systems 3(4), 323–350 (1992)CrossRefGoogle Scholar
  9. 9.
    Goh, S.L., Mandic, D.P.: An Augmented Extended Kalman Filter Algorithm for Complex-Valued Recurrent Neural Networks. Neural Computation 19(4), 1039–1055 (2007)zbMATHCrossRefGoogle Scholar
  10. 10.
    Granger, C.W.J., Ramanathan, R.: Improved methods of combining forecasts. Journal of Forecasting 3(2), 197–204 (1984)CrossRefGoogle Scholar
  11. 11.
    Hamilton, J.D.: Time series analysis. Princeton University Press, Princeton (1994)zbMATHGoogle Scholar
  12. 12.
    Harvey, A.C.: Forecasting, structural time series models and the Kalman filter. Cambridge University Press, Cambridge (1990)zbMATHGoogle Scholar
  13. 13.
    Hussain, A.J., Ghazali, R., Al-Jumeily, D., Merabti, M.: Dynamic Ridge Polynomial Neural Network for Financial Time Series Prediction. In: IEEE International Conference on Innovation in Information Technology, Dubai, pp. 1–5 (2006)Google Scholar
  14. 14.
    Ince, H., Trafalis, T.B.: A hybrid model for exchange rate prediction. Decision Support Systems 42(2), 1054–1062 (2006)CrossRefGoogle Scholar
  15. 15.
    Kaastra, I., Boyd, M.: Designing a Neural Network for Forecasting Financial and Economic Time Series. Neurocomputing 10(3), 215–236 (1996)CrossRefGoogle Scholar
  16. 16.
    Makridakis, S., Andersen, A., Carbone, R., Fildes, R., Hibon, M., Lewandowski, R., Newton, J., Parzen, E., Winkler, R.: The accuracy of extrapolation (time series) methods: Results of a forecasting competition. Journal of Forecasting 1(2), 111–153 (1982)CrossRefGoogle Scholar
  17. 17.
    Newbold, P., Granger, C.W.J.: Experience with Forecasting Univariate Time Series and the Combination of Forecasts. Journal of the Royal Statistical Society 137(2), 131–165 (1974)MathSciNetCrossRefGoogle Scholar
  18. 18.
    Palm, F.C., Zellner, A.: To combine or not to combine? issues of combining forecasts. Journal of Forecasting 11(8), 687–701 (1992)CrossRefGoogle Scholar
  19. 19.
    Shapiro, A.F.: A Hitchhiker’s guide to the techniques of adaptive nonlinear models. Insurance: Mathematics and Economics 26(2-3), 119–132 (2000)zbMATHCrossRefGoogle Scholar
  20. 20.
    Sermpinis, G., Laws, J., Dunis, C.L.: Modelling and trading the realised volatility of the FTSE100 futures with higher order neural networks. European Journal of Finance, 1–15 (2012)Google Scholar
  21. 21.
    Tenti, P.: Forecasting foreign exchange rates using recurrent neural networks. Applied Artificial Intelligence 10(6), 567–581 (1996)CrossRefGoogle Scholar
  22. 22.
    Terui, N., Van Dijk, H.K.: Combined forecasts from linear and nonlinear time series models. International Journal of Forecasting 18(3), 421–438 (2002)CrossRefGoogle Scholar
  23. 23.
    Yang, Y.: Combining Forecasting Procedures: Some theoretical results. Econometric Theory 20(1), 176–222 (2004)MathSciNetzbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Georgios Sermpinis
    • 1
  • Christian Dunis
    • 2
  • Jason Laws
    • 3
  • Charalampos Stasinakis
    • 1
  1. 1.Business SchoolUniversity of GlasgowUK
  2. 2.Liverpool Business SchoolUK
  3. 3.Management SchoolUniversity of LiverpoolUK

Personalised recommendations