Abstract
From Takens Theorem, we know that we can characterize the autonomous part of a dynamical system by a sequence of observations. If we want to force a neural net to learn this dynamics, we can use a sequence of time horizons as output. But in normal network structures these different outputs are learned nearly independently. By a new net architecture we allow additional information flows between the different outputs and as a sequence we get a better representation of the underlying dynamical system. The net is based on multilayer feed forward architecture. The model can then be converted to its equivalent nonlinear state space representation. Using this state space form, a Coupled Neural Net algorithm based on Extended Kalman Filter is derived to estimate the state. We analyze the net of a chaotic time series (Logistic map) by using the dynamical invariant that characterizes the attractor, the largest lyaponov exponent. A detailed step by step description of the methodology is presented to facilitate the use of this new method. The pertinence of this model is discussed from the Tunisian Stock Exchange database.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Blank, S.C.: Chaos in future Markets? a Nonlinear Dynamical Analysis. The Journal of Future Markets. 11.6 (1991) 711–728
Albano, A., Passamante, A., Hediger, T., Farell, M. E.: Using Neural Nets to look for chaos. Physica D. 58 (1992) 1
Nychka, D., Ellner, S., Gallant, A.R., Maccaffrey, D.: Findings chaos in noisy systems. Journal of the Royal Statistical Society, Series B. 54 (1992) 399–426
Hertz, J., Krogh, A., Palmer, R.: Introduction to the Theory of Neural Computing, Santa Fe Institute Studies in Science of Complexity, Amsterdam: Addison-Wesly (1991)
Weigend,.A.S., Huberman, B.A., Rumelhart, D.E.: Predicting the Future: A Connectionnist Approach. International Journal of Neural Systems, Vol 1.3 (1991) 193–209
Hornik, K., Stinchcombe, M., White: Multilayer feed forward networks are universal approximators. Neural Networks, vol.2 (1989) 359–366.
Aoki, M.: State-Space Modeling of Times Series, Berlin: Springer-Verlag (1987)
Wolf, A., Swift, J.B., Swinney, H.L., Vastano, J.A.: Determining Lyaponov Exponents from a time series. Physica D. 16 (1985) 285–317
Hsieh, D.A.: Chaos and Nonlinear Dynamics:Application to Financial Markets. The Journal of Finance. 6.5 (1991) 1839–1877
Hillmer, S.C, Trabelsi, A.: Benchmarking of Economic Time Series. Journal of the American Statistical Association: Theory and Methods. 82 (1987) 1064–1071
Trabelsi, A., Hillmer, S.C.: A Benchmarking Approach to Forecast Combination. Journal of Business and Economic Statistics. (1989) 353–362
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Slim, C., Trabelsi, A. (2003). Neural Network for Modeling Financial Time Series: A New Approach. In: Kumar, V., Gavrilova, M.L., Tan, C.J.K., L’Ecuyer, P. (eds) Computational Science and Its Applications — ICCSA 2003. ICCSA 2003. Lecture Notes in Computer Science, vol 2669. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-44842-X_25
Download citation
DOI: https://doi.org/10.1007/3-540-44842-X_25
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-40156-8
Online ISBN: 978-3-540-44842-6
eBook Packages: Springer Book Archive