Deep Learning Architectures for Predictive Modeling in Financial Time Series

Year : 2025 | Volume : 15 | Issue : 03 | Page : 45 55
    By

    Nagajayant Nagamani,

  1. Software Engagement, Virtusa, Chennai, Tamil Nadu, India

Abstract

This study investigates the application of deep learning architectures, particularly convolutional neural networks (CNNs), to the challenging task of financial time series forecasting. Financial markets are inherently complex and influenced by a range of factors, making accurate prediction of price movements a difficult problem. In this research, historical financial data including stock prices, volumes, and other relevant indicators are used to train CNN models aimed at capturing the underlying patterns and temporal dependencies in market behavior. The convolutional layers help in extracting latent features that traditional statistical methods might overlook, enabling the network to learn both local and global structures within the data. The performance of the models is evaluated through rigorous back testing on diverse datasets, demonstrating promising predictive capability that can be leveraged for improved financial decision-making and risk management. Additionally, the study identifies key challenges such as overfitting, non- stationarity of financial data, and the need for more sophisticated architectures or hybrid approaches. Recommendations are provided to guide future research towards refining model design, incorporating multi-modal data, and enhancing optimization techniques to achieve greater predictive accuracy and robustness. Overall, this work contributes to the growing body of knowledge in applying deep learning to finance, offering valuable insights for both academic researchers and industry practitioners.

Keywords: Reinforcement Learning, Intrinsic Rewards, Exploration Strategy, Random Latent Exploration(RLE), Latent Vector Conditioning , Latent Distribution , Adaptive Exploration Varients .

[This article belongs to Current Trends in Signal Processing ]

How to cite this article:
Nagajayant Nagamani. Deep Learning Architectures for Predictive Modeling in Financial Time Series. Current Trends in Signal Processing. 2025; 15(03):45-55.
How to cite this URL:
Nagajayant Nagamani. Deep Learning Architectures for Predictive Modeling in Financial Time Series. Current Trends in Signal Processing. 2025; 15(03):45-55. Available from: https://journals.stmjournals.com/ctsp/article=2025/view=222449


References

  1.  Cheng Y, Xu Z, Chen Y, Wang Y, Lin Z, Liu J. A deep learning framework integrating CNN and BiLSTM for financial systemic risk analysis and prediction. arXiv [Preprint]. 2025 Feb. Available from: arXiv:2502.06847
  2.  Duong TN, Le TQ, Do TTA, Nguyen MT, Phan TH, Nguyen TT. Investigating market strength prediction with CNNs on candlestick chart images. arXiv [Preprint]. 2025 Jan. Available from: arXiv:2501.12239
  3.  Wen H. Prediction of stock price by neural network based on CNN, LSTM, ANN. Adv Econ Manag Polit Sci. 2024;87:229–37.
  4.  Thaker A, Chan LH, Sonner D. Forecasting agriculture commodity futures prices with convolutional neural networks with application to wheat futures. J Risk Financ Manag. 2024 Apr;17(4):143.
  5. Xu Z. Research on the application of convolutional neural network in stock market forecasting. Adv Eng Technol Res. 2024 Apr;10(1):430–7.
  6.  Varshney S, Srivastava P. A comparative study of future stock price prediction through artificial neural network and ARIMA modelling. NMIMS Manag Rev. 2023 Dec;31(4):1–10.
  7. Zhou Z, Wu R. Stock price prediction model based on convolutional neural networks. J Ind Eng Appl Sci. 2024;2(4):1–10.
  8. Chahuán-Jiménez K. Neural network-based predictive models for stock market index forecasting. J Risk Financ Manag. 2024 Jun;17(6):242.
  9. Wang J. Stock price prediction using convolutional neural networks on various time frames. Highlights Sci Eng Technol. 2024;2(4):1–10.
  10. Liu Y, Zhang H, Li Q, Chen J, Wang R. Deep convolutional transformer network for stock movement prediction. Electronics. 2024 Oct;13(21):4225.
  11.  Jia D, Gao Q, Deng H. Stock market prediction based on time-frequency analysis and convolutional neural network. J Phys Conf Ser. 2022;2224(1):012017.
  12.  Xiang S, Zhang Y, Liu X, Wang H, Li J. Temporal and heterogeneous graph neural network for financial time series prediction. arXiv [Preprint]. 2023 May. Available from: arXiv:2305.08740
  13. Aadhitya A, Rajapriya R, Vineetha RS, Bagde MA. Predicting stock market time-series data using CNN-LSTM neural network model. arXiv [Preprint]. 2023 May. Available from: arXiv:2305.14378
  14. Lea C, Chan KH, Wong J, Ho KL. Increasing the Hong Kong stock market predictability: a temporal convolutional network approach. Comput Econ. 2024;61:1665–79.
  15.  Kumar S, Reddy V, Sharma P, Patel A, Mehta R. Stock market forecasting using a neural network through fundamental indicators, technical indicators and market sentiment analysis. Comput Econ. 2024;61:1234–45.
  16.  Zhao Z, Li Y, Sun H, Guo X. A model based LSTM and graph convolutional network for stock trend prediction. PeerJ Comput Sci. 2023;9:e2326.
  17. ovak , eluš ek redi tion of sto k pri e movement ased on dail high pri es uant Finance. 2023;16(5):793–826.
  18. Di Persio L, Honchar O. Recurrent neural networks approach to the financial forecast of Google assets. Int J Math Comput Simul. 2023;11:7–13.
  19.  LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2023;521(7553):436–44.
  20. Hager GD, Lea C, Vidal R, Reiter A. Temporal convolutional networks for action segmentation and detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2023. p. 156–65.
Regular Issue Subscription Original Research
Volume 15
Issue 03
Received 05/07/2025
Accepted 25/07/2025
Published 07/08/2025
Publication Time 33 Days
90


My IP

PlumX Metrics