Otimização de Portfólio Futuro Baseado em Aprendizagem Profunda e Algoritmo Evolutivo Multiobjetivo
Resumo
O mercado de ações é importante para o crescimento econômico, pois fornece a oportunidade para investidores formar patrimonio e fornece recursos para o crescimento de organizaçoes. Porém, o comportamento dinâmico do mercado é um desafio tanto para humanos quanto para algoritmos. Assim, este trabalho propoe um modelo regressivo usando uma rede neural profunda (LSTM) para encontrar padroes nos dados históricos de ações e tentar predizer os valores esperados para o futuro. Em seguida, faz-se a otimizaçao o portfólio utilizando um algoritmo evolutivo multiobjetivo chamado MOEA/D. Os dados utilizados nos testes sao dados históricos de dois indices conhecidos: o S&P 500 e o Dow Jones. Os resultados indicam que o modelo de prediçao associado ao MOEA/D pode sugerir portfolios com retorno entre 14% e 22% no indice S&P 500, e entre 4% e 10% no indice Dow Jones dependendo do risco que o investidor esta disposto a assumir.
Downloads
Referências
Arevalo, A., Nino, J., Hernandez, G., e Sandoval, J. (2016). High-frequency trading strategy based on deep neural networks. volume 9773, pages 424–436.
Bjorklund, S. e Uhlin, T. (2017). Artificial neural networks for financial time series prediction and portfolio optimization. Master’s thesis, Department of Management and Engineering, Linkoping University, Link ¨ oping, Sweden.
Bovespa (2000). O mercado de capitais: Sua importancia para o desenvolvimento e os entraves com que se defronta no brasil. Tendencia - Consultoria Integrada.
Campelo, F., Batista, L., e Aranha, C. (2018). The MOEADr package - a componentbased framework for multiobjective evolutionary algorithms based on decomposition. Journal of Statistical Software.
Chan, L., Karceski, J., e Lakonishok, J. (1999). On portfolio optimization: Forecasting covariances and choosing the risk model. Review of Financial Studies, 12:937–74.
Chen, T.-L. e Chen, F.-Y. (2016). An intelligent pattern recognition model for supporting investment decisions in stock market. Information Sciences, 346-347:261 – 274.
Chiang, W.-C., Enke, D., Wu, T., e Wang, R. (2016). An adaptive stock index trading decision support system. Expert Systems with Applications, 59:195 – 207.
Chong, E., Han, C., e Park, F. C. (2017). Deep learning networks for stock market analysis and prediction. Expert Syst. Appl., 83(C):187–205.
Chourmouziadis, K. e Chatzoglou, P. D. (2016). An intelligent short term stock trading fuzzy system for assisting investors in portfolio management. Expert Systems with Applications, 43:298 – 311.
Deb, K., Pratap, A., Agarwal, S., e Meyarivan, T. (2002). A fast and elitist multiobjective genetic algorithm: Nsga-ii. IEEE Transactions on Evolutionary Computation, 6(2):182–197.
Duffie, D. e Pan, J. (1997). An overview of value at risk. The Journal of Derivatives, 4(3):7–49.
Economatica (2015). Aumento da taxa dos pequenos investidos que entraram nas negociacões da bolsa.
Enke, D. e Mehdiyev, N. (2013). Stock market prediction using a combination of stepwise regression analysis, differential evolution-based fuzzy clustering, and a fuzzy inference neural network. Intelligent Automation and Soft Computing, 19.
Harper, D. R. (2019). An introduction to value at risk (var). https://www.investopedia.com/articles/04/092904.asp.
Heaton, J. B., Polson, N. G., e Witte, J. H. (2016). Deep learning for finance: deep portfolios: J. b. heaton, n. g. polson and j. h. witte. Applied Stochastic Models in Business and Industry, 33.
Hochreiter, S. e Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8):1735–1780.
Krokhmal, P., Palmquist, J., e Uryasev, S. (2003). Portfolio optimization with conditional value-at-risk objective and constraints. Journal of Risk, 4.
Morettin, P. e Toloi, C. M. C. (2011). Analise de séries temporais. Edgard Blucher, ed Brasil, 3 edition.
Niaki, S. T. A. e Hoseinzade, S. (2013). Forecasting s&p 500 index using artificial neural networks and design of experiments. Journal of Industrial Engineering International, 9.
Nugent, C. (2018). S&p 500 stock data.
Patel, J., Shah, S., Thakkar, P., e Kotecha, K. (2015). Predicting stock market index using fusion of machine learning techniques. Expert Systems with Applications, 42(4):2162–2172.
Patterson, J. e Gibson, A. (2017). Deep Learning A Practitioner’s Approach. O’Reilly Media, Inc., 1005 Gravenstein Highway North, Sebastopol, CA 95472.
Pereira, I. L., Crispim, S. F., e Savoia, J. R. F. (2013). Valor de mercado como indicador de competitividade: identificac¸ao de empresas brasileiras com alto desempenho no longo prazo. Revista das Faculdades Integradas Vianna Junior.
Qiu, M., Song, Y., e Akagi, F. (2016). Application of artificial neural network for the prediction of stock market returns: The case of the japanese stock market. Chaos, Solitons & Fractals, 85:1–7.
Sak, H., Senior, A., e Beaufays, F. (2014). Long short-term memory recurrent neural network architectures for large scale acoustic modeling. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, pages 338–342.
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., e Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15:1929–1958.
Weng, B. (2017). Application of machine learning techniques for stock market prediction. PhD thesis, Graduate Faculty of Auburn University, Auburn, Alabama.
Xiao, J., Li, J.-J., Hong, X.-X., Huang, M.-M., Hu, X.-M., Tang, Y., e Huang, C.-Q. (2018). An improved moea/d based on reference distance for software project portfolio optimization. Complexity, 2018:1–16.
Zhang, Q. e Li, H. (2008). Moea/d: A multiobjective evolutionary algorithm based on decomposition. Evolutionary Computation, IEEE Transactions on, 11:712 – 731.
Zhong, X. e Enke, D. (2017). Forecasting daily stock market return using dimensionality reduction. Expert Systems with Applications, 67:126 – 139.
