Back-propagation artificial neural networks in stock market forecasting. An application to the Warsaw Stock Exchange WIG20

• Autores: Yasen Rajihy, Kesra Nermend, Akeel Alsakaa
• Localización: Aestimatio: The IEB International Journal of Finance, ISSN 2173-0164, Nº. 15, 2017, págs. 88-99
• Idioma: inglés
• Títulos paralelos:
  • Redes neuronales artificialesretropropagación para la predicción en mercados de valores. Una Aplicación al índice de la Bolsa de Varsovia (WIG20)
• Enlaces
  • Texto completo (pdf)
• Resumen
  • español

    La predicción en los mercados de valores juega un papel crucial en la teoría y práctica de la inversión, sobre todo tras los progresos que han tenido lugar en el desarrollo de sistemas de negociación automatizados en los mercados de capitales. Las técnicas tradicionales, como el análisis estadístico, el análisis fundamental y el análisis técnico, ya no son consideradas como las mejores opciones en el terreno de la predicción financiera y las redes neuronales artificiales han ganado popularidad, sobre todo en ámbito de la predicción bursátil. El algoritmo de retropropagación es uno de los sistemas de entrenamiento de redes neuronales más populares cuando de predicción financiera se trata. Uno de los mayores problemas en lo que se refiere al uso de redes neuronales entrenadas con un algoritmo de retropropagación es la determinación tanto del número de capas ocultas como del número de neuronas en cada una de dichas capas. En este artículo se presentan tres arquitecturas de una red neuronal con conexiones hacia adelante (feed-forward) y con una única capa oculta que difieren en el número de neuronas que contiene dicha capa oculta. Estas tres redes, entrenadas con un algoritmo de retropropagación, compiten en cuanto a su capacidad para predecir el comportamiento del Índice de la Bolsa de Varsovia (WIG20, que incluye las veinte mayores compañías cotizadas en la Bolsa de Varsovia).

    Los resultados obtenidos muestran que la arquitectura consistente en una capa de entrada con N neuronas, una capa oculta con 3/2(N+1) neuronas y una capa de salida con una neurona es la que proporciona las mejores predicciones.

  • English

    Stock market forecasting plays a key role in investment practice and theory, especially given the progress made in developing automated trading systems for use in capital markets.

    Traditional techniques such as statistical analysis, fundamental analysis and technical analysis are no longer considered the best options in this field and artificial neural networks have gained in popularity, especially for use in stock market forecasting. The back-propagation algorithm is one of the most popular neural network training algorithms in financial time series prediction. One of the biggest problems regarding the use of neural networks trained with the back-propagation algorithm is the determination of both the number of hidden layers and the number of neurons included in each hidden layer. In this article, we present three competing architectures of a feed-forward network, all with only one hidden layer but differing in the number of neurons included in that layer. The three models will be trained with the back-propagation algorithm, in order to determine which one provides the best forecasting performance. Data on the WIG20 (Capitalization-Weighted Stock Market Index of the 20 largest companies on the Warsaw Stock Exchange) are used to evaluate the performance of the competing architectures. The results obtained show that the architecture consisting of an input layer with N neurons, one hidden layer with 3/2(N+1) neurons and an output layer with one neuron outperforms the other two models

• Referencias bibliográficas
  • Adebiyi, A.A., Ayo, C.K., Adebiyi, M.O. and Otokiti, S.O. (2012). Stock price prediction using neural network with hybridized market indicators,...
  • Aiken M. and Bsat, M. (1999). ‘Forecasting Market Trends with Neural Networks, Information Systems Management, 16, pp. 42-46.
  • Asadi, S., Hadavandi, E., Mehmanpazir, F. and Nakhostin, M.M. (2012). Hybridization of evolutionary Levenberg–Marquardt neural networks and...
  • Atsalakis, G.S. and Valavanis, K.P. (2009). Surveying stock market forecasting techniques–Part II: Soft computing methods, Expert Systems...
  • Azoff, E.M. (1994). Neural Network Time Series Forecasting of Financial Market, Wiley, New York.
  • Badaway, F.A., Abdelazim, H.Y. and Darwish, M.G. (2006). Genetic Algorithms for Predicting the Egyptian Stock Market, Proceedings of the2005...
  • Balachandher, K.G., Lai, C.H., Fauzias, M.N. and Nor, F.M. (2002). An examination of the random walk model and technical trading rules in...
  • Benvenuto, N. and Piazza, F. (1992). On the complex backpropagation algorithm. IEEE Transactions on Signal Processing, 40(4), pp. 967-969.
  • Boyacioglu, M.A. and Avci, D. (2010). An adaptive network-based fuzzy inference system (ANFIS) for the prediction of stock market return:...
  • Cao, L. (2003). Support vector machines experts for time series forecasting, Neurocomputing, 51, pp. 321-339.
  • Cilimkovic, M. (2011). Neural Networks and Back Propagation Algorithm, Institute of Technology Blanchardstown, Blanchardstown Road North Dublin,...
  • Chan, M.C., Wong, C.C. and Lam, C.C. (2000). Financial time series forecasting by neural network using conjugate gradient learning algorithm...
  • Choudhry, R. and Garg, R. (2008). A hybrid machine learning system for stock market forecasting, International Journal of Computer, Electrical,...
  • Duke, J.A. (1993). Neural network futures trading—a feasibility study. In Adaptive Intelligent Systems, Society for Worldwide Interbank Financial...
  • Kaastra, I. and Boyd, M. (1996). Designing a Neural Network for Forecasting Financial and Economic Time Series, Neurocomputing, 10, pp. 215-236.
  • Kohara, K., Ishikawa, T., Fukuhara, Y. and Nakamura, Y. (1997). Stock price prediction using prior knowledge and neural networks, International...
  • Garliauskas, A. (1999). Neural network chaos and computational algorithms of forecast in finance, Proceedings of the IEEE International Conference...
  • Guresen, E., Kayakutlu, G. and Daim, T.U. (2011). Using artificial neural network models in stock market index prediction, Expert Systems...
  • Hall, J.W. (1994). Adaptive selection of U.S. stocks with neural nets, in G.J. Deboeck (Ed.), Trading on the Edge: Neural, Genetic and Fuzzy...
  • Han, Q. (2002). Theory, Design and Application of Artificial Neural Network. Chemical Industry Press, New Jersey.
  • Heaton, J. (2008). Introduction to neural networks with Java. 2nd Edition. Heaton Research Inc.
  • Hiemstra, Y. (1995). Modeling structured nonlinear knowledge to predict stock market returns, in R.R. Trippi (Ed.), Chaos & nonlinear...
  • Hwarng, H.B. (2001). Insights into Neural Networks Forecasting of Time Series corresponding to ARMA (p,q) Structures, International Journal...
  • Kao L.-J., Chiu, C.-C., Lu, C.-J. and Yang, J.-L.(2013). Integration of nonlinear independent component analysis and support vector regression...
  • Kimoto, T. Asakawa, K., Yoda, M. and Takeoka, M. (1990). Stock market prediction system with modular neural networks. Neural Networks 1990...
  • Lam, M. (2004). Neural network techniques for financial performance prediction: integrating fundamental and technical analysis, Decision Support...
  • Lee T.S. and Chiu, C.C. (2002). Neural network forecasting of an opening cash price index, International Journal of Systems Science, 33, pp....
  • Lee T.S. and Chen, N.J. (2002). Investigating the information content of non-cash-trading index futures using neural networks, Expert Systems...
  • Lee, C.J. (2009). Financial time series forecasting using independent component analysis and support vector regression, Decision Support Systems,...
  • Leigh W., Purbis, R. and Ragusa, J.M. (2002). Forecasting the NYSE composite index with technical analysis, pattern recognizer, neural network,...
  • Leigh, W., Hightower, R. and Modani, N. (2005). Forecasting the New York stock exchange composite index with past price and interest rate...
  • Liu, H. and Wang, J. (2011). Integrating Independent Component Analysis and Principal Component Analysis with Neural Network to Predict Chinese...
  • Lu, C.J. (2010). Integrating independent component analysis-based denoising scheme with neural network for stock price prediction, Expert...
  • Mantri, J.K., Gahan, P. and Nayak, B.B. (2014). Artificial neural networks—an application to stock market volatility. In J.K. Mantri (Ed.),...
  • Nikolopoulos, C. and Fellrath, P.F. (1994). A hybrid expert system for investment advising, Expert Systems, 11(4), pp. 245-250.
  • Romahi, Y. and Shen, Q. (2000). Dynamic Financial Forecasting with Automatically Induced Fuzzy Associations, IEEE, pp. 493-498.
  • Rumelhart, Y.C. (1995). Backpropagation: Theory, architectures, and applications, Lawrence Erlbaum Associates, New Jersey.
  • Sun, Z.L., Choi, T.M., Au, K.F. and Yu, Y. (2008). Sales forecasting using extreme learning machine with applications in fashion retailing,...
  • Thawornwong, D.E. (2005). The use of data mining and neural networks for forecasting stock market returns, Expert Systems with Applications,...
  • Ticknor, J.L. (2013). A Bayesian regularized artificial neural network for stock market forecasting, Expert Systems with Applications, 40(14),...
  • Trippi, R.R. and DeSieno, D. (1992). Trading equity index futures with a neural network, The Journal of Portfolio Management, 19(1), pp. 27-33.
  • Werbos, J.P. (1988). Generalization of back propagation with application to a recurrent gas market model, Neural Networks, 1, pp. 339-356.
  • White, H. (1989). Learning in artificial neural networks: a statistical perspective, Neural Computation, 1, pp. 425-464.