Procedimiento para la previsión de tipos de cambio. Aplicación de redes neuronales artificiales EUR/USD
DOI:
https://doi.org/10.33936/eca_sinergia.v13i2.4149Palavras-chave:
inteligencia artificial, pronósticos, redes neuronales recurrentes, series de tiempo.Resumo
A previsão da taxa de câmbio é um elemento fundamental na estratégia de hedge contra o risco cambial, o que impulsiona a busca por novas técnicas mais eficazes. A fim de facilitar o uso de algoritmos de inteligência artificial para este fim, foi criado o procedimento apresentado neste trabalho. Este procedimento baseia-se na utilização de métodos quantitativos para obtenção da previsão, sendo o mais importante o uso de redes neurais artificiais, especificamente aquelas chamadas de long-short term memory. É realizada a construção de diferentes modelos, a partir dos quais se escolhe o ótimo usando a raiz quadrada do erro quadrático médio como critério de seleção. Após a obtenção de modelos que executam com eficácia compreendida na faixa de 97,28% - 99,55%, a eficácia do procedimento foi demonstrada.Palavras chave: inteligência artificial, previsões, redes neurais recorrentes, séries temporais
Downloads
Não há dados estatísticos.
Referências
Allaire, J. J., & Chollet, F. (2020). keras: R Interface to 'Keras'. R package version 2.3.0.0. https://CRAN.R-project.org/package=keras
Allaire, J. J., & Tang,Y. (2020). tensorflow: R Interface to 'TensorFlow'. R package version 2.2.0. Recuperado de: https://CRAN.R-project.org/package=tensorflow
Brownlee, J. (2016). «Time series forecasting as Supervised Learning». Machine Learning Mastery. Recuperado de de: https://machinelearningmastery.com/time-series-forecasting-supervised-learning/
Brownlee, J. (2017). «How to use timesteps in LSTM networks for time series forecasting». Machine Learning Mastery. Reccuperado de: https://machinelearningmastery.com/use-timesteps-lstm-networks-time-series-forecasting/
Brownlee, J. (2018). «Difference Between a Batch and an Epoch in a Neural Network». Machine Learning Mastery. Recuperado de: https://machinelearningmastery.com/difference-between-a-batch-and-an-epoch/
Brownlee, J. (2019). «The promise of recurrent neural networks for time series forecasting». Machine Learning Mastery. Recuperado de: https://machinelearningmastery.com/promise-recurrent-neural-networks-time-series-forecasting/
Chollet, F., & Allaire, J. J. (2018). Deep Learning with R. Manning Publication.
De la Oliva, F.,& García, A.(2020). «Procedimiento para un pronostico de la tasa de cambio euro-dólar». Revista CoFin Habana, 1. Recuperado de: http://www.cofinhab.uh.cu/index.php/RCCF/articledownload/387/377
Duncan, E.W. (2020). «A Guide to (Re-)Installing R and Related Software». Rpubs. Recuperado de: https://rpubs.com/Earlien/guide-to-installing-R
Hirekerur, R. (2020). «A Comprehensive Guide To Loss Functions – Part1:Regression». Medium. Obtenido de: https://medium.com/analytics-vidhya/a-comprehensive-guide-to-loss-functions-part-1-regression-ff8b847675d6
Hyndman, R., Athanasopoulos, G., Bergmeir, C., Caceres, G., Chhay, L., O'Hara-Wild, M., Petropoulos, F., Razbash, S., Wang, E., & Yasmeen, F. (2020). _forecast: Forecasting functions for time series and linear models_. R package version 8.12. Recuperado: http://pkg.robjhyndman.com/forecast.
Hyndman, RJ., & Khandakar ,Y. (2008). “Automatic time series forecasting: the forecast package for R.” _Journal of Statistical Software_, *26*(3), 1-22. Recuperado de: http://www.jstatsoft.org/article/view/v027i03.
Keydana, S. (23 de mayo de 2017). RPubs. «Time series forecasting - with deep learning». Rpubs. Recuperado de: https://rpubs.com/zkajdan/279967
Olah, C. (27 de agosto de 2015). «Understanding LSTM networks». Colah's blog. Recuperado de: https://colah.github.io/posts/2015-08-Understanding-LSTMs/
Phillips, N.D (2018). YaRrr! The Pirate’s Guide to R. Bookdown. Se encuentra en: Recuperado de https://bookdown.org/ndphillips/YaRrr/
R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Recuperado de https://www.R-project.org/
RStudio Team (2019). RStudio: Integrated Development for R. RStudio, Inc., Boston, MA. Recuperado de http://www.rstudio.com/.
Torres, J. (2018). «Learning process of a neural network». Medium. Se encuentra en: https://towardsdatascience.com/how-to-artificial-neural-networks-learn-773e46399fc7
Trapletti, T., & Hornik,K. (2019). tseries: Time Series Analysis and Computational Finance. R package version 0.10-47. Recuperado de https://CRAN.R-project.org/package=tseries
Wickham, H., & Seidel,D. (2020). scales: Scale Functions for Visualization. R package version 1.1.1. Recuperado de https://CRAN.R-project.org/package=scales
Allaire, J. J., & Tang,Y. (2020). tensorflow: R Interface to 'TensorFlow'. R package version 2.2.0. Recuperado de: https://CRAN.R-project.org/package=tensorflow
Brownlee, J. (2016). «Time series forecasting as Supervised Learning». Machine Learning Mastery. Recuperado de de: https://machinelearningmastery.com/time-series-forecasting-supervised-learning/
Brownlee, J. (2017). «How to use timesteps in LSTM networks for time series forecasting». Machine Learning Mastery. Reccuperado de: https://machinelearningmastery.com/use-timesteps-lstm-networks-time-series-forecasting/
Brownlee, J. (2018). «Difference Between a Batch and an Epoch in a Neural Network». Machine Learning Mastery. Recuperado de: https://machinelearningmastery.com/difference-between-a-batch-and-an-epoch/
Brownlee, J. (2019). «The promise of recurrent neural networks for time series forecasting». Machine Learning Mastery. Recuperado de: https://machinelearningmastery.com/promise-recurrent-neural-networks-time-series-forecasting/
Chollet, F., & Allaire, J. J. (2018). Deep Learning with R. Manning Publication.
De la Oliva, F.,& García, A.(2020). «Procedimiento para un pronostico de la tasa de cambio euro-dólar». Revista CoFin Habana, 1. Recuperado de: http://www.cofinhab.uh.cu/index.php/RCCF/articledownload/387/377
Duncan, E.W. (2020). «A Guide to (Re-)Installing R and Related Software». Rpubs. Recuperado de: https://rpubs.com/Earlien/guide-to-installing-R
Hirekerur, R. (2020). «A Comprehensive Guide To Loss Functions – Part1:Regression». Medium. Obtenido de: https://medium.com/analytics-vidhya/a-comprehensive-guide-to-loss-functions-part-1-regression-ff8b847675d6
Hyndman, R., Athanasopoulos, G., Bergmeir, C., Caceres, G., Chhay, L., O'Hara-Wild, M., Petropoulos, F., Razbash, S., Wang, E., & Yasmeen, F. (2020). _forecast: Forecasting functions for time series and linear models_. R package version 8.12. Recuperado: http://pkg.robjhyndman.com/forecast.
Hyndman, RJ., & Khandakar ,Y. (2008). “Automatic time series forecasting: the forecast package for R.” _Journal of Statistical Software_, *26*(3), 1-22. Recuperado de: http://www.jstatsoft.org/article/view/v027i03.
Keydana, S. (23 de mayo de 2017). RPubs. «Time series forecasting - with deep learning». Rpubs. Recuperado de: https://rpubs.com/zkajdan/279967
Olah, C. (27 de agosto de 2015). «Understanding LSTM networks». Colah's blog. Recuperado de: https://colah.github.io/posts/2015-08-Understanding-LSTMs/
Phillips, N.D (2018). YaRrr! The Pirate’s Guide to R. Bookdown. Se encuentra en: Recuperado de https://bookdown.org/ndphillips/YaRrr/
R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Recuperado de https://www.R-project.org/
RStudio Team (2019). RStudio: Integrated Development for R. RStudio, Inc., Boston, MA. Recuperado de http://www.rstudio.com/.
Torres, J. (2018). «Learning process of a neural network». Medium. Se encuentra en: https://towardsdatascience.com/how-to-artificial-neural-networks-learn-773e46399fc7
Trapletti, T., & Hornik,K. (2019). tseries: Time Series Analysis and Computational Finance. R package version 0.10-47. Recuperado de https://CRAN.R-project.org/package=tseries
Wickham, H., & Seidel,D. (2020). scales: Scale Functions for Visualization. R package version 1.1.1. Recuperado de https://CRAN.R-project.org/package=scales
Downloads
Publicado
2022-05-20
Como Citar
Oliva de Con, F. de la, Molina Fernández, R., & Díaz Rodríguez, D. (2022). Procedimiento para la previsión de tipos de cambio. Aplicación de redes neuronales artificiales EUR/USD. ECA Sinergia, 13(2), 107–117. https://doi.org/10.33936/eca_sinergia.v13i2.4149
Edição
Seção
Artículos
Licença
Copyright (c) 2022 Fidel Oliva
Este trabalho está licenciado sob uma licença Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
