Journal Press India^®

Author Details ( * ) denotes Corresponding author

Purpose: This paper applies neural network models to predict the daily returns of the BSE (Bombay Stock Exchange) Sensex. Multilayer perceptron network is used to build the daily returns model and the network is trained using the Error Back Propagation algorithm. Design/Methodology/Approach: The data consists of daily index values of the BSE Sensex. The period under consideration is 16/01/1980 to 26/09/1997. The data set consists of 3667 data points. The data has been obtained from the Capitaline 2000 database that provides daily stock market data. The entire analysis has been done basically on the daily returns rather than the raw index value. Findings: The results show that the predictive power of the network model is influenced by the previous day’s return than the first three-day’s inputs. The study shows that satisfactory results can be achieved when applying neural networks to predict the BSE Sensex. Research Limitations: Further, experimentation is required for producing better prediction of stock prices and further work has to be done by testing it for weekly or monthly returns, as well as by including other micro and macro-economic variables as inputs. Managerial Implications: Global financial market players, institutional investors and generic software developers should consider developing stock market trading strategies using neural networks. Originality/Values: An attempt is made in this study to understand the use of neural networks in the field of finance.

Keywords

Stock market prediction, Neural networks, Financial forecasting, nonlinear time series analysis

1. Altman, E., M.Giancarlo, and F.Varentto (1994), “Corporate Distress Diagnosis: Comparisons using Linear Discriminant Analysis and Neural Networks (the Italian Experience), Journal of Banking and Finance, Vol.18, pp.505-29.
2. Burgess, A.N. “Statistical Yield Curve Arbitrage in Eurodollar Futures using Neural Networks”, Department of Computer Science, London Business School, London, U.K.
3. Burgess, A.N. & Refenes, A.N. (1996), “The use of Error feedback terms in Neural Network Modeling of Financial Time Series”, in Dunic, C.(ed.) Quantitative Methods in Trading and Finance, Wiley & Sons.
4. Collins, E., Ghosh, S. and Svofield. C. (1998), “An Application of a multiple Neural Network Learning System to Emulation of Mortgage Underwriting Judgements”, Proceedings of the IEEE International Conference on Neural Networks, Vol.2, pp.459-466.
5. Dutta, S., and Shekhar, S. (1998), “Bond Rating: A Non-conservative Application of Neural Networks”, Proceedings of the IEEE International Conference on Neural networks, Vol.2, pp.443-450.
6. Hatfield, Gay B., and White, A. Jay (1996), “Bond Rating Changes: Neural Net Estimates for Bank Holding Companies”, Working Paper.
7. Hutchinson, James M. Andrew W.Lo and Tomaso Poggio (1994), “A Non-Parametric Approach to Pricing and Hedging Derivatives Securities Via Learning Networks” – The Journal of Finance, Vol. XLIX, No.3, July.
8. Kusturelis, Jason E. (1998) “Forecasting Financial Markets using Neural Networks: An Analysis of Methods and Accuracy”, Master of Science in Management Thesis, Department of System Management, http://www.nps.navy.mil/Code36/ KUTSURE.html
9. McCluskey, Peter C. (1993) “Feed-forward and Recurrent Neural Networks and Genetic Programs for Stock Market and Time Series Forecasting”, Department of Computer Science, Brown University, September.
10. Refenes, A.N. and Zapranis, A.D. “Neural Networks in Tactical Asset Allocation: A Comparative Study with Regression Models”, Department of Decision Science, London Business School, London.
11. Refenes , A.N., Bentz, Y. Bunn, D.W., Burgess, A.N. and Zapranis, A.D. (1997) “Financial Time Series Modeling with Discounted Least Squares back Propagation”, Neuro Computing Vol. 14, pp.123-138.
12. Suan, Tan Sen and Koh Hian Chye (1997), “Neural Network Applications in Accounting and Business”, Accounting and Business Review, Vol.4, No.2, July, pp.297-317.