A Quantitative Analysis Decision System Based on Deep Learning and NSGA-II for FX Portfolio Prediction

  • Hua Shen
  • Xun Liang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10868)


Forecasting foreign exchange (FX) rate and optimizing FX portfolio with the help of Artificial Intelligence has aroused wide interest among global capital market. As far as we know, this is the first paper which, from the perspective of institutional and individual investors, proposes a complete quantitative analysis decision system based on Deep Learning and NSGA-II to forecast FX rate and select FX portfolio successively. To be specific, we provide a whole procedure from data collection to FX forecasting with Stacked Autoencoders and further to optimal FX portfolio selection with NSGA-II. Furthermore, an empirical analysis has been conducted with 28 FX currency pairs, in which our algorithm has been compared with two other machine learning algorithms. Ultimately, our system provides optimized FX portfolio solutions for investors with diverse preference.


Deep learning Stacked autoencoders NSGA-II 



The work was supported by the National Natural Science Foundation of China (No. 71531012), and the Natural Science Foundation of Beijing (No. 4172032).


  1. 1.
    Deng, S., Yoshiyama, K., Mitsubuchi, T., Sakurai, A.: Hybrid method of multiple kernel learning and genetic algorithm for forecasting short-term foreign exchange rates. Comput. Econ. 45, 49–89 (2015)CrossRefGoogle Scholar
  2. 2.
    Shen, F., Chao, J., Zhao, J.: Forecasting exchange rate using deep belief networks and conjugate gradient method. Neurocomputing 167, 243–253 (2015)CrossRefGoogle Scholar
  3. 3.
    Zhang, R., Shen, F., Zhao, J.: A model with fuzzy granulation and deep belief networks for exchange rate forecasting. In: International Joint Conference on Neural Networks, pp. 366–373 (2014)Google Scholar
  4. 4.
    Shen, H., Liang, X.: A time series forecasting model based on deep learning integrated algorithm with stacked autoencoders and SVR for FX prediction. In: Villa, A.E.P., Masulli, P. (eds.) ICANN 2016. LNCS, vol. 9887, pp. 326–335. Springer, Cham (2016). Scholar
  5. 5.
    Li, H., Zhang, Q.: Multiobjective optimization problems with complicated Pareto sets, MOEA/D and NSGA-II. IEEE Trans. Evol. Comput. 13, 284–302 (2009)CrossRefGoogle Scholar
  6. 6.
    Kannan, S., Baskar, S., Mccalley, J.D., Murugan, P.: Application of NSGA-II algorithm to generation expansion planning. IEEE Trans. Power Syst. 24, 454–461 (2009)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.RenMin University of ChinaBeijingChina

Personalised recommendations