Investigation of Prediction Capabilities using RNN Ensembles

Nijolė Maknickienė, Algirdas Maknickas


Modern portfolio theory of investment-based financial market forecasting use probability distributions. This investigation used a neural network architecture, which allows to obtain distribution for predictions. Comparison of the two different models - points based prediction and distributions based prediction - opens new investment opportunities. Dependence of forecasting accuracy on the number of EVOLINO recurrent neural networks (RNN) ensemble was obtained for five forecasting points ahead. This study allows to optimize the computational time and resources required for sufficiently accurate prediction.


  1. Brezak, D., Bacek, T., Majetic, D., Kasac, J., and Novakovic, B. (March, 2012). A comparison of feedforward and recurrent neural networks in time series forecasting. In Computational Intelligence for Financial Engineering Economics (CIFEr), 2012 IEEE Conference on, pages 1-6.
  2. Dalcin, L. (12/01/2012).
  3. Fox, G., Johnson, M., Lyzenga, G., Otto, S., Salmon, J., and Walker, D. (1988). Solving problems on concurrent processors vol 1. 1373849.
  4. Garcia-Pedrajas, N., Hervas-Martinez, C., and Ortiz-Boyer, D. (June, 2005). Cooperative coevolution of artificial neural network ensembles for pattern classification. Evolutionary Computation, IEEE Transactions on, 9(3):271-302.
  5. Hutter, M. (2007). On universal prediction and bayesian confirmation. Theoretical Computer Science, 384(1):33-48.
  6. Kaastra, I. and Boyd, M. (1996). Designing a neural network for forecasting financial and economic time series. Neurocomputing, 10(3):215-236.
  7. Kumar, V., Gmma, A., and Anshul, G. (1994). Introduction to parallel computing: Design and analysis of algorithms.
  8. Maknickas, A. and Maknickiene, N. (2012). Influence of data orthogonality to accuracy and stability of financial market predictions. In The Fourth International Conference on Neural Computation Theory and Applications (NCTA 2012), pages 616-619, Barcelona, Spain.
  9. Nguyen, H. and Chan, C. (2004). Multiple neural networks for a long term time series forecast. Neural Computing & Applications, 13(1):90-98.
  10. Riley, B. (2012). Practical statistical inference for the opinions a.of biased experts blake riley. In In: The 50th Annual Meeting of the MVEA, volume 2. Missouri Valley Economic Association.
  11. Rutkauskas, A. and Lapinskait-Vvohlfahrt, I. (2010). Marketing finance strategy based on effective risk management. In The Sixth International Scientific Conference Business and Managemen, volume 1, pages 162-169. Technika.
  12. Rutkauskas, A., Miec?inskien, A., and Stasytyt, V. (2008). Investment decisions modelling along sustainable development concept on financial markets. Technological and Economic Development of Economy, 14(3):417-427.
  13. Rutkauskas, A. V. (2000). Formation of adequate investment portfolio for stochasticity of profit possibilities. Property management, 4(2):100-115.
  14. Rutkauskas, A. V. (2012). Using sustainability engineering to gain universal sustainability efficiency. Sustainability, 4(6):1135-1153.
  15. Rutkauskas, A. V. and Stasytyte, V. (2011). Optimal portfolio search using efficient surface and threedimensional utility function. Technological and Economic Development of Economy, 17(2):305-326.
  16. Schmidhuber, J., Wierstra, D., Gagliolo, M., and Gomez, F. (2007). Training recurrent networks by evolino. Neural Computation, 19(3):757-779.
  17. Schmidhuber, J., Wierstra, D., and Gomez, F. (2005). Modeling systems with internal state using evolino. In In Proc. of the 2005 conference on genetic and evolutionary computation (GECCO), pages 1795-1802, Washington. ACM Press, New York, NY, USA.
  18. Siwek, K., Osowski, S., and Szupiluk, R. (2009). Ensemble neural network approach for accurate load forecasting in a power system. International Journal of Applied Mathematics and Computer Science, 19(2):303-315.
  19. Tsakonas, A. and Dounias, G. (2005). An architecturealtering and training methodology for neural logic networks: Application in the banking sector. In Madani, K., editor, Proceedings of the 1st International Workshop on Artificial Neural Networks and Intelligent Information Processing, ANNIIP 2005, pages 82-93, Barcelona, Spain. NSTICC Press. In conjunction with ICINCO 2005.
  20. Uchigaki, S., Uchida, S., Toda, K., and Monden, A. (Aug.2012). An ensemble approach of simple regression models to cross-project fault prediction. In Software Engineering, Artificial Intelligence, Networking and Parallel Distributed Computing (SNPD), 2012 13th ACIS International Conference on, pages 476- 481.
  21. Walczak, S. (2001). An empirical analysis of data requirements for financial forecasting with neural networks. Journal of management information systems, 17(4):203-222.
  22. Wierstra, D., Gomez, F. J., and Schmidhuber, J. (2005). Evolino: Hybrid neuroevolution / optimal linear search for sequence learning. In Proceedings of the 19th International Joint Conference on Artificial Intelligence (IJCAI),, pages 853-858, Edinburgh.
  23. Zhou, Z.-H., Wu, J., and Tang, W. (2002). Ensembling neural networks: many could be better than all. Artificial intelligence, 137(1):239-263.

Paper Citation

in Harvard Style

Maknickienė N. and Maknickas A. (2013). Investigation of Prediction Capabilities using RNN Ensembles . In Proceedings of the 5th International Joint Conference on Computational Intelligence - Volume 1: NCTA, (IJCCI 2013) ISBN 978-989-8565-77-8, pages 391-395. DOI: 10.5220/0004554703910395

in Bibtex Style

author={Nijolė Maknickienė and Algirdas Maknickas},
title={Investigation of Prediction Capabilities using RNN Ensembles},
booktitle={Proceedings of the 5th International Joint Conference on Computational Intelligence - Volume 1: NCTA, (IJCCI 2013)},

in EndNote Style

JO - Proceedings of the 5th International Joint Conference on Computational Intelligence - Volume 1: NCTA, (IJCCI 2013)
TI - Investigation of Prediction Capabilities using RNN Ensembles
SN - 978-989-8565-77-8
AU - Maknickienė N.
AU - Maknickas A.
PY - 2013
SP - 391
EP - 395
DO - 10.5220/0004554703910395