Abstract
There is growing interest both in the field of neural computing and in the financial world in the possibility of using neural networks to forecast the future changes in prices of stocks, exchange rates, commodities and other financial time series. Since networks have been shown to be capable of modelling the underlying structure of a time series, many attempts have been made at exploiting that capability in order to carry out a technical analysis of such prices. If the efficient markets hypothesis is true, however, there is no underlying structure to be modelled, and the whole endeavour is doomed to failure. This paper investigates the common methods for such an approach, and outlines the major pitfalls and common errors to avoid. The author hopes that by pointing out the possible pitfalls now, we can avoid making claims to the commercial world before we are properly ready to do so.
Similar content being viewed by others
References
White H. Economic prediction using neural networks: the case of ibm daily stock returns. In: Proceedings IEEE International Conference on Neural Networks, San Diego 1988; II. 451–459
Lowe D, Webb AR. Time series prediction by adaptive networks: a dynamical systems perspective. In: Proceedings IEEE 1991; 28(1): 17–24
Waibel A. Modular construction of time delay neural networks for speech recognition. Neural Computation 1989; 1: 39–46
Refenes AN. Constructive learning and its application to currency exchange rate forecasting. In: Neural Network Applications in Investment and Finance Services, E Turban, R Trippi (eds), Probus Publishing, 1991
Refenes AN, Azema-Barac M. Neural network applications in financial asset management. Journal of Neurocomputing and Applications 1994; 2(1): 13–39
Huberman BA, Weigend AS, Rummelhart DE. Predicting sunspots and exchange rates with connectionist networks. In: Nonlinear Modeling and Forecasting, SFI Studies in the Science of Complexity Proc. Vol. XII, M Casdagli, S Eubank (eds), 1992; 395–432
Mozer MC. Neural net architectures for temporal sequence processing. In: Time Series Prediction. Forecasting the Future and Understanding the Past, AS Weigend, NA Gershenfeld (eds), Addison Wesley, 1993; 243–264
Jiang B, Jang G, Lai F, Pan C. An intelligent stock portfolio management system based on short-term trend prediction using dual-module neural networks. In: Artificial Neural Networks, O Simula, T Kohonen, K Makisara, J Kangas (eds), Elsevier, 1991; 447–452
Elman JL. Finding structure in time. Cognitive Science 1990; 14: 179–211
Williams RJ, Zipser D. A learning algorithm for continually running fully recurrent neural networks. Neural Computation 1989; 1: 270–280
Cleeremans A. Finite state automata and simple recurrent networks. Neural Computation 1989; 2(1): 372–381
Swingler KM. Sequence categorisation using multiple recurrent layers to create many trajectories through network state space. In: M Marinara, G Morasso (eds), Proceedings ICANN, Vol. 2, 1994: 1025–1028. Springer-Verlag
Lin L, Mitchell TM. Memory approaches to reinforcement learning in non-markovian domains. CMU Technical report CMU-CS-92-188, 1992
Psarrou A, Buxton H. Motion analysis with recurrent neural nets. In: M Marinara, G Morasso (eds), Proceedings ICANN 1994; 1: 54–57. Springer-Verlag
Kendall MG, Stuart A. The Advanced Theory of Statistics, Vol. 3, Charles Griffin & Co., 1968
Weigend AS, Gershenfeld NA. Time Series Prediction. Forecasting the Future and Understanding the Past. Addison Wesley, 1993
Zhang X, Hutchison J. Simple architectures on fast machines: Practical issues in nonlinear time series prediction. In: Time Series Prediction. Forecasting the Future and Understanding the Past, AS Weigend, NA Gershenfeld (eds), Addison Wesley, 1993; 219–241
DTI. Best Practice Guidelines for developing Neural Computing Applications. Touche Ross for the DTI, 1994
LeBaron B. Non linear diagnostics and simple trading rules for high frequency foreign exchange rates. In: Time Series Prediction. Forecasting the Future and Understanding the Past, AS Weigend, NA Gershenfeld (eds), Addison Wesley, 1993; 457–474
Pettit RR. Dividend announcements, security performance and capital market efficiency. Journal of Finance 1972: 993–1007
Hoptroff RG. The principles and practice of time series forecasting and business modelling using neural nets. Neural Computing and Applications 1993; 1: 59–66
Tan KY, Kiang MY. Managerial applications of neural networks: The case of bank failure predictions. Management Science 1992; 38(7): 926–947
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Swingler, K. Financial prediction: Some pointers, pitfalls and common errors. Neural Comput & Applic 4, 192–197 (1996). https://doi.org/10.1007/BF01413817
Issue Date:
DOI: https://doi.org/10.1007/BF01413817