A Pi-Sigma Higher Order Neural Network for Stock Index Forecasting

Conference paper
Part of the Smart Innovation, Systems and Technologies book series (SIST, volume 32)

Abstract

Multilayer perceptron (MLP) has been found to be most frequently used model for stock market forecasting. MLP is characterized with black-box in nature and lack of providing a formal method of deriving ultimate structure of the model. Higher order neural network (HONN) has the ability to expand the input representation space, perform high learning capabilities that require less memory in terms of weights and nodes and have been utilized in many complex data mining problems. To capture the extreme volatility, nonlinearity and uncertainty associated with stock data, this paper considered a HONN, called Pi-Sigma Neural Network (PSNN), for prediction of closing prices of five real stock markets. The tunable weights are optimized by Gradient Descent (GD) and a global search technique, Genetic Algorithm (GA). The model proves its superiority when trained with GA in terms of Average Percentage of Errors (APE).

Keywords

Stock index forecasting Multilayer perceptron Higher order neural network Pi-sigma neural network Genetic algorithm 

References

  1. 1.
    Guler, M., Sahin, E.: A new higher-order binary-input neural unit: learning and generalizing effectively via using minimal number of monomials. In: Third Turkish Symposium on Artificial Intelligence and Neural Networks Proceedings, 51–60. Middle East Technical University, Ankara, Turkey (1994)Google Scholar
  2. 2.
    Wang, Z., Fang, J., and Liu, X.: Global stability of stochastic high-order neural networks with discrete and distributed delays, Chaos, Solutions and Fractals, (2006). doi: 10.1016/j.chaos.2006, 06.063
  3. 3.
    Shin, Y., Ghosh, J.: Ridge polynomial networks. IEEE Trans. Neural Netw. 6, 610–622 (1995)CrossRefGoogle Scholar
  4. 4.
    Park, S., Smith, M.J.T., Mersereau, R.M.: Target recognition based on directional filter banks and higher-order neural networks. Dig. Sign. Process. 10, 297–308 (2000)Google Scholar
  5. 5.
    Leerink, L.R., Giles, C.L., Horne, B.G., Jabri, M.A.: Learning with product units. In: Tesaro, G., Touretzky, D., Leen, T. (eds.) Advances in neural information processing systems 7, pp. 537–544. MIT Press, Cambridge (1995)Google Scholar
  6. 6.
    Kwon, Y.K., Moon, B.R.: A hybrid neuro-genetic approach for stock forecasting. IEEE Trans. Neural Netw 18(3), 851–864 (2007)CrossRefMathSciNetGoogle Scholar
  7. 7.
    Yu, L., Zhang, Y. Q.: Evolutionary fuzzy neural networks for hybrid financial prediction. IEEE Trans. Syst Man Cyber Part C Appl Rev 35(2), 244–249 (2005)Google Scholar
  8. 8.
    Shin, Y., Ghosh, J.: The pi–sigma network: an efficient higher-order neural network for pattern classification and function approximation. In: International Joint Conference on Neural Networks (1991)Google Scholar
  9. 9.
    Perantonis, S., Ampazis, N., Varoufakis, S., Antoniou, G.: Constrained learning in neural networks: application to stable factorization of 2-d polynomials. Neural Process. Lett. 7(1), 5–14 (1998)CrossRefGoogle Scholar
  10. 10.
    Huang, D.S., Ip, H.H.S., Law, K.C.K., Chi, Z.: Zeroing polynomials using modified constrained neural network approach. IEEE Trans. Neural Netw. 16(3), 721–732 (2005)CrossRefGoogle Scholar
  11. 11.
    Shin, Y., Ghosh, J.: Efficient higher-order neural networks for classification and function approximation. Int. J. Neural Syst. 3, 323–350 (1992)CrossRefGoogle Scholar
  12. 12.
    Epitropakis, M.G., Plagianakos, V.P., Vrahatis, M.N.: Hardware-friendly higher-order neural network training using distributed evolutionary algorithms. Appl. Soft Comput. 10, 398–408 (2010)CrossRefGoogle Scholar
  13. 13.
    Ghazali, R., Husaini, N.H., Ismail, L.H., Samsuddin, N.A.: An Application of jordan pi-sigma neural network for the prediction of temperature time series signal. Recurrent Neural Networks and Soft Computing, (2012) Dr. Mahmoud ElHefnawi (Ed.), ISBN: 978-953-51-0409-4Google Scholar
  14. 14.
    Ghazali, R., Hussain, A.J., Liatsis, P.: Dynamic ridge polynomial neural network: forecasting the univariate non-stationary and stationary trading signals. Expert Syst. Appl. 38, 3765–3776 (2011). (Elsevier)CrossRefGoogle Scholar
  15. 15.
    Shin Y., Ghosh, J.: Realization of boolean functions using binary pi-sigma networks. In: Dagli, C.H., Kumara, S.R.T., Shin, Y.C. (eds.) Intelligent engineering systems through artificial neural networks, pp. 205–210. ASME Press, New York (1991)Google Scholar

Copyright information

© Springer India 2015

Authors and Affiliations

  1. 1.Veer Surendra Sai University of TechnologyBurlaIndia
  2. 2.Silicon Institute of TechnologyBhubaneswarIndia

Personalised recommendations