ANFIS Application for Time Series Forecasting
DOI:
https://doi.org/10.21501/21454086.927Keywords:
Forecasting, Time Series, Uncertainty, ANFIS Networks, Neuro-Fuzzy Nets, Stock Market Analysis.Abstract
This paper shows a Neuro-Fuzzy methodology which is applied to the financial problem of stock market forecasting in the short time, whose results can be used as a reference for speculative investments in Colombian stock exchange market as a complement to technical and fundamental analysis. Moreover, the development of an ANFIS (Adaptive Neuro-Based Fuzzy Inference System) MATLAB language based tool is presented. Such tool is founded on a heuristic combination of both neural networks and fuzzy logic, with consideration on the number and type of membership functions for input variables. The decision maker can trust in the model reliability by means of the calculation given by the error residual method. Some other measures are also defined to verify reliability of the predictions, including the average square error and the standard deviation that are directly calculated from the model itself.Downloads
References
A. Arango Londoño, “Pronóstico del Índice General de la Bolsa de Valores de Colombia (IGBC) usando modelos de inferencia difusa,” MSc Administración Tesis Maestría en Administración, Facultad de Minas, Universidad Nacional de Colombia, Medellín, 2013.
A. Arango, J. D. Velásquez, and C. J. Franco, “Técnicas de lógica difusa en la predicción de índices de mercados de valores: una revisión de literatura,” Revista Ingenierías Universidad de Medellín, vol. 12, pp. 117-126, 2013.
E. M. Toro Ocampo, A. Molina Cabrera, and A. Garcés Ruiz, “Pronóstico de bolsa de valores empleando técnicas inteligentes,” Revista Tecnura, vol. 9, pp. 57-66, 2006.
Z. Yun, Z. Quan, S. Caixin, L. Shaolan, L. Yuming, and S. Yang, “RBF neural network and ANFIS-based short-term load forecasting approach in real-time price environment,” Power Systems, IEEE Transactions on, vol. 23, pp. 853-858, 2008.
P. A. Sánchez, “Cambios estructurales en series de tiempo: una revisión del estado del arte,” Revista Ingenierías Universidad de Medellín, vol. 7, pp. 115-140, 2008.
M. Mathworks. (2013). Training routine for Sugeno-type Fuzzy Inference System. [Online]. Available: http://www.mathworks.es/es/help/fuzzy/anfis.html
M. Mathworks. (2013). Adaptive Neuro-Fuzzy Modeling. [Online]. Available: http://www.mathworks.es/es/help/fuzzy/adaptive-neuro-fuzzy-inference-systems.html
M. Mathworks. (2013). Fuzzy Inference System Modeling. [Online]. Available: http://www.mathworks.es/es/help/fuzzy/mamdani-fuzzy-inference-systems.html
M. Lam, “Neural network techniques for financial performance prediction: integrating fundamental and technical analysis,” Decision Support Systems, vol. 37, pp. 567-581, 9// 2004.
Y.-H. Lui and D. Mole, “The use of fundamental and technical analyses by foreign exchange dealers: Hong Kong evidence,” Journal of International Money and Finance, vol. 17, pp. 535-545, 6/1/ 1998.
N. Ülkü and E. Prodan, “Drivers of technical trend-following rules profitability in world stock markets,” International Review of Financial Analysis, vol. 30, pp. 214-229, 12// 2013.
D. A. Agudelo and J. H. Uribe, “¿Realidad o sofisma? Poniendo a prueba el análisis técnico en las acciones colombianas,” Cuadernos de Administración, vol. 22, pp. 189 -217, 2009.
R. Babuška and H. Verbruggen, “Neuro-fuzzy methods for nonlinear system identification,” Annual reviews in control, vol. 27, pp. 73-85, 2003.
K. Cios and W. Pedrycz, “Neuro-fuzzy algorithms,” Handbook of Neural Computation, IOP Publishing and Oxford University Press, Oxford, p. D1, 1997.
J.-S. Jang, “ANFIS: adaptive-network-based fuzzy inference system,” Systems, Man and Cybernetics, IEEE Transactions on, vol. 23, pp. 665-685, 1993.
M. Roubens, “Fuzzy sets and decision analysis,” Fuzzy sets and systems, vol. 90, pp. 199-206, 1997.
T. Takagi and M. Sugeno, “Fuzzy identification of systems and its applications to modeling and control,” Systems, Man and Cybernetics, IEEE Transactions on, pp. 116-132, 1985.
T. Whalen and C. Brønn, “Essentials of Decision Making Under Generalized Uncertainty,” in Combining Fuzzy Imprecision with Probabilistic Uncertainty in Decision Making. vol. 310, J. Kacprzyk and M. Fedrizzi, Eds., ed: Springer Berlin Heidelberg, 1988, pp. 26-47.
G. J. Correa Henao, “Aproximaciones metodológicas para la toma de decisiones, apoyadas en modelos difusos,” Máster Ingeniería de Sistemas Tesis Magister en Ingeniería de Sistemas Universidad Nacional de Colombia, Facultad de Minas, Universidad Nacional de Colombia, Sede Medellín, 2004.
J. F. Marín Valencia and K. A. Muñoz Ocampo, “Comparación entre los Modelos de Box & Jenkins y el Modelo ANFIS en el pronóstico de precios de acciones en el corto plazo,” Ingeniería Industrial Tesis Ingeniero Industrial, Facultad de Minas, Universidad Nacional de Colombia, Sede Medellín, 2002.
M. A. Henao, W. M. Trejos, and F. V. Duque, “Pronóstico de las tasas de cambio. Una aplicación al Yen Japonés mediante redes neuronales artificiales,” Scientia Et Technica, vol. 12, pp. 233-238, 2006.
T. Cruz, E. Arturo, J. H. Restrepo, and P. Medina Varela, “Pronóstico del índice general de la Bolsa de Valores de Colombia usando redes neuronales,” Scientia et Technica, vol. 1, 2009.
D. A. Agudelo Rueda, “Liquidez en los mercados accionarios colombianos,” Cuadernos de Administración, vol. 23, pp. 239-269, 2010.
S. Botero Botero and J. A. Cano Cano, “Análisis de series de tiempo para la predicción de los precios de la energía en la Bolsa de Colombia,” Cuadernos de Economía, vol. 27, pp. 173-208, septiembre, 2008.
M.-G. Financial. (2013). S&P Dow Jones Indices. [Online]. Available: http://us.spindices.com/
G. A. Marín, “Predicción de precios de acciones en bolsa mediante redes neuronales artificiales,” Especialización en Sistemas de Información, Trabajo de grado Especialista en Sistemas Informáticos, Facultad de Minas, Universidad Nacional de Colombia, Sede Medellín, 2000.
T. B. Ludermir, A. Yamazaki, and C. Zanchettin, “An Optimization Methodology for Neural Network Weights and Architectures,” Neural Networks, IEEE Transactions on, vol. 17, pp. 1452-1459, // 2006.
D. A. Agudelo Rueda, “Costos de transacción asociados a la liquidez en la Bolsa de Valores de Colombia,” Cuadernos de Administración, vol. 24, pp. 13-37, 2011.
B. V. C. BVC. (2013). Comunicados de Prensa Bolsa de Valores de Colombia. [Online]. Available: http://www.bvc.com.co/
A. Zúñiga and C. Jordán, “Pronóstico de caudales medios mensuales empleando sistemas Neurofuzzy,” Revista Tecnológica-ESPOL, vol. 18, 2013.
L. Á. Medina, “Aplicación de la teoría del portafolio en el mercado accionario colombiano,” Cuadernos de Economía, vol. 22, pp. 129-168, 2003.
A. Abraham and B. Nath, “A neuro-fuzzy approach for modelling electricity demand in Victoria,” Applied Soft Computing, vol. 1, pp. 127-138, 2001.
L. dos Santos Coelho, A. A. Portela Santos, and N. C. Affonso da Costa Jr., “Podemos prever a taxa de cambio brasileira? Evidência empírica utilizando inteligência computacional e modelos econométricos,” Gestão & Produção, vol. 15, pp 635-647, 2008.
D. Quintana, R. Gimeno, and P. Isasi, “Detección de inercia sectorial en salidas a bolsa mediante modelos ARIMA y redes neuronales,” Instituto Centroamericano de Administración Pública - ICAP, 2005.
F. Villada, E. García, and J. D. Molina, “Pronóstico del precio de la energía eléctrica usando redes neuro-difusas,” Información tecnológica, vol. 22, pp. 111-120, 2011.
Y. Gao and M. J. Er, “NARMAX time series model prediction: feedforward and recurrent fuzzy neural network approaches,” Fuzzy sets and systems, vol. 150, pp. 331-350, 2005.
A. E. Gaweda, J. M. Zurada, and R. Setiono, “Input selection in data-driven fuzzy modeling,” in Fuzzy Systems, 2001. The 10th IEEE International Conference on, 2001, pp. 1251-1254.
J.-S. Jang, “Input selection for ANFIS learning,” in Fuzzy Systems, 1996, Proceedings of the Fifth IEEE International Conference on, 1996, pp. 1493-1499.
J. D. Velásquez Henao, “Pronóstico de la serie de Mackey–Glass usando modelos de regresión no lineal,” Dyna, vol. 71, pp. 85-95, 2004.
J. D. Velásquez Henao and C. Zapata, “Pronóstico del caudal medio mensual, con una ventana de 12 meses, usando sistemas difusos,” 2004.
J. J. Montaño Moreno, “Redes neuronales artificiales aplicadas al análisis de datos,” Doctorado Psicología, Facultad de Psicología, Tesis doctoral de la Facultad de Psicología, Universitat de Les Illes Balears, Palma de Mallorca, España, 2002.
I. L. López-Cruz and L. Hernández-Larragoiti, “Modelos neuro-difusos para temperatura y humedad del aire en invernaderos tipo cenital y capilla en el centro de México,” Agrociencia, vol. 44, pp. 791-805, 2010.
P. Sinčák, M. Holécy, and M. Dučai. (1998). Computational Intelligence in Financial Cybernetics. [Online]. Available: http://www.ai-cit.sk/cigOld/source/publications/conference_papers/sincak/AII/html/index.html
S. Areerachakul and S. Sanguansintukul, “A comparison between the multiple linear regression model and neural networks for biochemical oxygen demand estimations,” in Eighth International Symposium on Natural Language Processing, 2009. SNLP '09, Bangkok, Thailand, 2009, pp. 11-14.
W. Yu and X. Li, “Fuzzy identification using fuzzy neural networks with stable learning algorithms,” Fuzzy Systems, IEEE Transactions on, vol. 12, pp. 411-420, 2004.
J. D. Velásquez and C. Zapata, “Pronóstico del caudal medio mensual, con una ventana de 12 meses, usando sistemas difusos,” Proyecto de Investigación. Universidad Nacional De Colombia–Medellín, 2004.
H. D. Álvarez Zapata, “Versión breve de algunas técnicas de Inteligencia Artificial,” Dyna, vol. 63, pp. 13-23, 1996.
L. P. Maguire, B. Roche, T. M. McGinnity, and L. McDaid, “Predicting a chaotic time series using a fuzzy neural network,” Information Sciences, vol. 112, pp. 125-136, 1998.
E. C. Zapata, J. D. Velásquez, and R. Smith Q, “Caracterización del SOI usando ANFIS con residuales heterocedásticos,” Ingeniare. Revista chilena de ingeniería, vol. 15, pp. 302-312, 2007.
P. Estévez García, “Aplicaciones de las redes neuronales en finanzas,” Documentos de trabajo de la Facultad de Ciencias Económicas y Empresariales, p. 5, 2002.
N. Science. (2004). ANNI Standard - Artificial Neural Network Investing. [Online]. Available: http://download.cnet.com/ANNI-Standard-Artificial-Neural-Network-Investing/3000-2057_4-10343970.html?tag=pdp_prod
Downloads
Published
How to Cite
Issue
Section
License
In accordance with national and international copyrights, as well as publishing policies of "Fundación Universitaria Luis Amigó" and its Journal "Lámpsakos" (indexed with ISSN : 2145-4086), I (we ) hereby manifest:1. The desire to participate as writers and submit to the rules established by the magazine publishers.
2. The commitment not to withdraw the manuscript until the journal finishes the editing process of the ongoing issue.
3. That article is original and unpublished and has not been nominated or submitted together in another magazine; therefore, the rights of the article in evaluation have not been assigned in advance and they do not weigh any lien or limitation for use.
4. The absence of conflict of interest with commercial institution or association of any kind
5. The incorporation of the quotes and references from other authors, tending to avoid plagiarism. Accordingly, the author affirms that the paper being published do not violate copyright, intellectual property or privacy rights of third parties. Morover, if necessary there is a way of demonstrating the respective permits original copyright to the aspects or elements taken from other documents such as texts of more than 500 words, tables, graphs, among others. In the event of any claim or action by a third party regarding copyright on the article, the author (s) will assume full responsibility and come out in defense of the rights herein assigned. Therefore, for all purposes, the Journal "Lámpsakos" of the "Fundación Universitaria Luis Amigó" acts as a third party in good faith.
6. In the event of the publication of the article, the authors free of charge and on an exclusive basis the integrity of the economic rights and the right to print, reprint and reproduction in any form and medium, without any limitation as to territory is concerned, in favor of the Journal "Lámpsakos" of the "Fundación Universitaria Luis Amigó".
