Predicting stock market behavior by combining multilayer perceptron neural networks and dynamic mode decomposition

Document Type : Research Paper

Authors

Department of Mathematics, Faculty of Basic Sciences, Payame Noor University, Tehran, Iran

10.22108/msci.2025.142940.1697

Abstract

he stock market acts as an indicator of the overall health of the economy, and its proper performance indicates growth in businesses and the expansion of the economy. In this research, based on Dynamic Mode Decomposition (DMD) and Multilayer Perceptron (MLP) neural network, a new hybrid method is presented for stock market forecasting. This hybrid method (DMD-MLP) extracts dominant and consistent data and uses them to predict the upward or downward trend of stock prices. To prove the effectiveness of this method, examples of different groups of the stock market have been presented, where in them, there are bullish, bearish, or neutral behaviors. These examples include Iran Gelatin Capsule Production, Ofogh Kourosh, Day Bank, Mellat Bank, Dana Insurance, and Pasargad Bank. The results show that the proposed method performs better than the MLP neural network in predicting the movements of the stock market, and the use of the DMD algorithm in MLP has a significant effect on improving predictions.

Keywords

Main Subjects

References

[1] R. Achkar, F. Elias-Sleiman, H. Ezzidine and N. Haidar, Comparison of BPA-MLP and LSTM-RNN for stocks prediction, in 6th International symposium on computational and business intelligence (ISCBI), (2018) 48–51.
[2] G. Ahmadi and Z. Akbari, Investigating air pollutants in Lorestan province and predicting their concentration using multi-layer neural network with stable online training (case studies: Khorramabad and Poldakhter), Nivar, 48 no. 126-127 (2024) 109–126.
[3] G. Ahmadi and M. Dehghandar, Chaotic time series prediction using rough-neural networks, Math. Interdisc. Res., 8 no. 2 (2023) 71–92.
[4] C. C. Aggarwal, Neural networks and deep learning, Springer, 2018.
[5] C. M. Bishop and N. M. Nasrabadi, Pattern recognition and machine learning, Springer, 2006.
[6] M. Doaei, S. A. Mirzaei and M. Rafigh, Hybrid multilayer perceptron neural network with grey wolf optimization for predicting stock market index, Advances in mathematical finance and applications, 6 no. 4 (2021) 883–894.
[7] I. Goodfellow, Y. Bengio and A. Courville, Deep learning, MIT Press, 2016.
[8] S. Haykin, Neural networks and learning machines (3rd ed.), Pearson, 2009.
[9] A. Kurani, P. Doshi, A. Vakharia, et al. A Comprehensive comparative study of artificial neural network
(ANN) and support vector machines (SVM) on stock forecasting, Ann. Data. Sci., 10 (2023) 183–208.
[10] Y. C. Liang, H. P. Lee, S. P. Lim, W. Z. Lin, K. H. Lee and C. G. Wu, Proper orthogonal decomposition and its applications-part i: Theory, Journal of Sound and Vibration, 252 no. 3 (2002) 527–544.
[11] R. Karimkhani, Y. E. Tabriz and G. Ahmadi, A new hybrid method of dynamic mode decomposition and long short-term memory for financial market forecasting, Journal of Mathematics and Modeling in Finance, 3 no. 2 (2023) 1–16.
[12] B. O. Koopman, Hamiltonian systems and transformation in Hilbert space, Proceedings of the National Academy of Sciences, 17 no. 5 (1931) 315–318.
[13] J. N. Kutz, S. L. Brunton, B. W. Brunton and J. L. Proctor, Dynamic mode decomposition: data-driven modeling of complex systems, SIAM, 2016.
[14] M. Lu and X. Xu, TRNN: An efficient time-series recurrent neural network for stock price prediction, Information Sciences, 657 (2024) 119951.
[15] M. L. Minsky and S. A. Papert, Perceptrons: an introduction to computational geometry, MIT Press, Cambridge, 1968.
[16] J. Mann and J. N. Kutz, Dynamic mode decomposition for financial trading strategies, Quantitative Finance, 16 no. 11 (2016) 1643–1655.
[17] T. M. Mitchell, Machine learning (McGraw-Hill international editions computer science series), 1997.
[18] M. P. Naeini, H. Taremian and H. B. Hashemi, Stock market value prediction using neural networks, in International conference on computer information systems and industrial management applications (CISIM), (2010) 132–136.
[19] A. T. Oyewole, O. B. Adeoye, W. Afua Addy, C. C. Okoye, O. C. Ofodile and C. E. Ugochukwu, Predicting stock market movements using neural networks: A review and application study, Computer Science and IT Research Journal, 5 no. 3 (2024) 651–670.
[20] T. Peters, Data-driven science and engineering: machine learning, dynamical systems, and control, Cambridge, Cambridge University Press, 2019.
[21] F. Rosenblatt, The perceptron, a perceiving and recognizing automaton Project Para, Cornell Aeronautical Laboratory, 1957-58.
[22] P. J. Schmid, Dynamic mode decomposition of numerical and experimental data, Journal of fluid mechanics, 656 (2010) 5–28.
[23] Y. G. Song, Y. L. Zhou and R. J. Han, Neural networks for stock price prediction, arXiv preprint arXiv:1805.11317, (2018) pp. 13.
[24] G. W. Stewart, On the early history of the singular value decomposition, SIAM Review, 35 no. 4 (1993) 551–566.
[25] R. S. Sutton and A. G. Barto, Reinforcement learning: an introduction (2nd ed.), MIT Press, 2018.
[26] J. H. Tu, Dynamic mode decomposition: Theory and applications, Princeton University, 2013.
[27] D. Wu, X. Ma and D. L. Olson, Financial distress prediction using integrated Z-score and multilayer perceptron neural networks, Decision Support Systems, 159 (2022) 113814.
[28] P. Yu and X. Yan, Stock price prediction based on deep neural networks, Neural Computing and Applications, 32 no. 6 (2020) 1609–1628.

Files

History

  • Receive Date: 01 October 2024
  • Revise Date: 30 May 2025
  • Accept Date: 08 June 2025
  • Publish Date: 21 October 2025

Share

How to cite

Statistics