Avaliação dos diferentes tipos de redes LSTM para predição de ações na bolsa de valores

  • Gabriel Souto CEFET/RJ
  • Bruna Capistrano CEFET/RJ
  • Matheus Matias CEFET/RJ
  • Jorge Soares CEFET/RJ
  • Eduardo Ogasawara CEFET/RJ
  • Lucas Giusti CEFET/RJ

Resumo


Redes Neurais Profundas são modelos valiosos na tarefa de aprendizagem. Neste trabalho, propomos o uso do método multicamadas conhecido como Long Short-Term Memory. Aplicamos três modelos diferentes (LSTM Vanilla, Stacked e Convolucional) para a mesma série de ações. Essa escolha foi feita devido à lacuna na literatura ao comparar quais modelos LSTM podem ser usados na predição de séries temporais. Os resultados encontrados comprovaram-se uma alternativa ao que pretendemos mostrar, no sentido de um trabalho comparativo com os melhores modelos LSTM.
Palavras-chave: Redes Neurais Profundas, Redes LSTM, Predição de Ações

Referências

B.M. Barber, Y.-T. Lee, Y.-J. Liu, and T. Odean. The cross-section of speculator skill: Evidence from day trading. Journal of Financial Markets, 18(1):1–24, 2014.

O. Bustos and A. Pomares-Quimbaya. Stock market movement forecast: A Systematic review. Expert Systems with Applications, 156, 2020.

T. Fischer and C. Krauss. Deep learning with long short-term memory networks for financial market predictions. European Journal of Operational Research, 270(2):654–669, 2018.

S. Hochreiter and J. Schmidhuber. Long Short-Term Memory. Neural Computation, 9(8):1735–1780, 1997.

Rob J. Hyndman and George Athanasopoulos. Forecasting: principles and practice. OTexts, May 2018. ISBN 978-0-9875071-1-2.

F. Kurniawan, D.E. Herwindiati, and M.D. Lauro. Raw Paper Material Stock Forecasting with Long Short-Term Memory. In 2021 9th International Conference on Information and Communication Technology, ICoICT 2021, pages 342–347, 2021.

B. Lindemann, T. Müller, H. Vietz, N. Jazdi, and M. Weyrich. A survey on long short-term memory networks for time series prediction. In Procedia CIRP, volume 99, pages 650–655, 2021.

D.M.Q. Nelson, A.C.M. Pereira, and R.A. De Oliveira. Stock market’s price movement prediction with LSTM neural networks. In Proceedings of the International Joint Conference on Neural Networks, volume 2017-May, pages 1419–1426, 2017.

M. Obthong, N. Tantisantiwong, W. Jeamwatthanachai, and G. Wills. A survey on machine learning for stock price prediction: Algorithms and techniques. In FEMIB 2020 - Proceedings of the 2nd International Conference on Finance, Economics, Management and IT Business, pages 63–71, 2020.

E.W. Saad, D.V. Prokhorov, and D.C. Wunsch II. Comparative study of stock trend prediction using time delay, recurrent and probabilistic neural networks. IEEE Transactions on Neural Networks, 9(6):1456–1470, 1998.

S. Selvin, R. Vinayakumar, E.A. Gopalakrishnan, V.K. Menon, and K.P. Soman. Stock price prediction using LSTM, RNN and CNN-sliding window model. In 2017 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2017, volume 2017-January, pages 1643–1647, 2017.

X. Shi, Z. Chen, H. Wang, D.-Y. Yeung, W.-K. Wong, and W.-C. Woo. Convolutional LSTM network: A machine learning approach for precipitation nowcasting. In Advances in Neural Information Processing Systems, volume 2015-January, pages 802–810, 2015.

S. Siami-Namini, N. Tavakoli, and A. Siami Namin. A Comparison of ARIMA and LSTM in Forecasting Time Series. In Proceedings - 17th IEEE International Conference on Machine Learning and Applications, ICMLA 2018, pages 1394–1401, 2019.

Ruey S. Tsay. Analysis of Financial Time Series. Wiley, July 2010. ISBN 978-0-470-64455-3.

Y. Wu, M. Yuan, S. Dong, L. Lin, and Y. Liu. Remaining useful life estimation of engineered systems using vanilla LSTM neural networks. Neurocomputing, 275:167–179, 2018.
Publicado
23/11/2021
SOUTO, Gabriel; CAPISTRANO, Bruna; MATIAS, Matheus; SOARES, Jorge; OGASAWARA, Eduardo; GIUSTI, Lucas. Avaliação dos diferentes tipos de redes LSTM para predição de ações na bolsa de valores. In: ESCOLA REGIONAL DE INFORMÁTICA DO RIO DE JANEIRO (ERI-RJ), 4. , 2021, Online. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2021 . p. 65-71. DOI: https://doi.org/10.5753/eri-rj.2021.18776.