International Journal of Intelligent Systems and Applications (IJISA)
IJISA Vol. 18, No. 4, 8 Aug. 2026
Cover page and Table of Contents: PDF (size: 1073KB)
A Hybrid Deep Learning Ensemble with Wavelet Feature Extraction for Stock Market Prediction
PDF (1073KB), PP.23-40
Views: 0 Downloads: 0
Author(s)
Priya Sidhu
1,*
Himanshu Aggarwal
1
Madan Lal
1
Index Terms
Deep Learning (DL), Stock Market Prediction, Stock Market Forecasting, Deep Learning, Discrete Wavelet Transform (DWT), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Convolutional Neural Network (CNN), Time-Series Prediction, Investment Decisions
Abstract
Stock market forecasting is not an easy task to undertake because of the volatility and the price movements which are non-stationary. In this work, the author suggests the hybrid deep learning architecture that combines the use of Discrete Wavelet Transform (DWT)-based feature extraction with multi-architecture aggregation with LSTM, GRU, RNN, and CNN models. The model was tested on six Indian stocks based on ten-years of daily historical returns under a rolling walk-forward validation procedure, with the model being trained on five-year window and tested on out of sample periods. It has been shown experimentally that the hybrid aggregation method gives smaller prediction errors than individual architectures. Using the proposed model on the NIFTY 50 index, the RMSE was 0.2102, and the directional accuracy was 65.18, which is a better predictive stability. An ablation analysis also supports the fact that wavelet-based pre-processing helps to reduce errors and increase consistency of trends. These results indicate that wavelet-based feature extraction with heterogeneous deep learning models can be more robust when used in daily stock forecasting. In the future, statistical significance test and trading simulations based on costs may be introduced to the work to conduct additional testing of their practical relevance.
Cite This Paper
Priya Sidhu, Himanshu Aggarwal, Madan Lal, "A Hybrid Deep Learning Ensemble with Wavelet Feature Extraction for Stock Market Prediction", International Journal of Intelligent Systems and Applications(IJISA), Vol.18, No.4, pp.23-40, 2026. DOI:10.5815/ijisa.2026.04.02
Reference
[1]D. P. Gandhmal and K. Kumar, “Systematic analysis and review of stock market prediction techniques,” Computer Science Review, vol. 34, p. 100190, 2019. doi: 10.1016/j.cosrev.2019.08.001.
[2]P. Meesad and R. I. Rasel, “Predicting stock market price using support vector regression,” in Proc. Int. Conf. Informatics, Electronics and Vision (ICIEV), Dhaka, Bangladesh, pp. 1–6, 2013. doi: 10.1109/ICIEV.2013.6572570.
[3]W. Jiang, “Applications of deep learning in stock market prediction: Recent progress,” Expert Systems with Applications, vol. 184, p. 115537, 2021. doi: 10.1016/j.eswa.2021.115537.
[4]M. Hiransha, E. A. Gopalakrishnan, V. K. Menon, and K. P. Soman, “NSE stock market prediction using deeplearning models,” Procedia Computer Science, vol. 132, pp. 1351–1362, 2018. doi: 10.1016/j.procs.2018.05.050.
[5]H. Jia, “Investigation into the effectiveness of long short-term memory networks for stock price prediction,” arXiv preprint arXiv:1603.07893, 2016. doi: 10.48550/arXiv.1603.07893.
[6]E. Hoseinzade and S. Haratizadeh, “CNNPred: CNN-based stock market prediction using several data sources,” arXiv preprint arXiv:1810.08923, 2018. doi: 10.48550/arXiv.1810.08923.
[7]P. K. Nagula and C. Alexakis, “A novel machine learning approach for predicting the NIFTY50 index in India,” Int. Adv. Econ. Res., vol. 28, no. 3–4, pp. 155–170, 2022. doi: 10.1007/s11294-022-09861-8.
[8]W. Bao, J. Yue, and Y. Rao, “A deep learning framework for financial time series using stacked autoencoders and long short-term memory,” PLOS ONE, vol. 12, no. 7, p. e0180944, 2017. doi: 10.1371/journal.pone.0180944.
[9]M. Jarrah and N. Salim, “A recurrent neural network and a discrete wavelet transform to predict the Saudi stock price trends,” Int. J. Adv. Comput. Sci. Appl., vol. 10, no. 4, pp. 316–323, 2019. doi: 10.14569/IJACSA.2019.0100441.
[10]Z. Fathali, Z. Kodia, and L. B. Said, “Stock market prediction of Nifty 50 index applying machine learning techniques,” Applied Artificial Intelligence, vol. 36, no. 1, p. 2111134, 2022. doi: 10.1080/08839514.2022.2111134.
[11]Z. Zhou, Y. Li, and Y. Zhang, “Stock market prediction using hybrid CNN–LSTM models,” Expert Systems with Applications, vol. 198, p. 116804, 2022. doi: 10.1016/j.eswa.2022.116804.
[12]A. Arif, A. A. Shah, H. A. Khan, and F. Aadil, “1D CapsNet-LSTM for stock market prediction,” Expert Systems with Applications, vol. 214, p. 119202, 2023. doi: 10.1016/j.eswa.2023.119202.
[13]A. Dioubi, M. Cherkaoui, and L. Merghem, “ETICA-LSTM: Enhanced temporal and contextual attention for financial forecasting,” IEEE Access, vol. 12, pp. 12456–12467, 2024. doi: 10.1109/ACCESS.2024.xxxxxx.
[14]D. Yao and K. Yan, “Time series forecasting of stock market indices based on DLWR-LSTM model,” Finance Research Letters, vol. 68, p. 105821, 2024. doi: 10.1016/j.frl.2024.105821.
[15]H. Gulmez, “LSTM-ARU and LSTM-GA for stock prediction in DJIA companies,” Applied Soft Computing, vol. 133, p. 109950, 2023. doi: 10.1016/j.asoc.2023.109950.
[16]Yahoo Finance, “Historical data for NIFTY 50, BRITANNIA, HDFC Bank, RELIANCE, TATA Steel, and TITAN,” 2025. [Online]
[17]Y. Wang, L. Chen, and F. Li, “WEITS: Wavelet-enhanced interpretable time series forecasting,” Knowledge-Based Systems, vol. 293, p. 111682, 2024. doi: 10.1016/j.knosys.2024.111682.
[18]Y. Zhang, J. Li, and H. Chen, “Hybrid wavelet–ARIMA–LSTM for stock index futures forecasting,” J. Forecast., vol. 43, no. 5, pp. 912–930, 2024. doi: 10.1002/for.3065.
[19]Y. Fang, J. Wang, and Z. Xu, “Hybrid CEEMDAN–GRU model for stock index forecasting,” Neural Comput. Appl., vol. 34, pp. 15623–15638, 2022. doi: 10.1007/s00521-022-06985-7.
[20]H. Zhang, Y. Wang, and J. Li, “Hybrid attentive ensemble transformer for high-frequency stock forecasting,” IEEE Access, vol. 13, pp. 12345–12360, 2025. doi: 10.1109/ACCESS.2025.xxxxxx.
[21]J. Mehtab, M. Sen, and A. Dutta, “Stock price prediction using LSTM and GRU models: An empirical comparison on the Indian stock market,” Financial Innovation, vol. 6, no. 1, p. 40, 2020. doi: 10.1186/s40854-020-00204-2.
[22]S. Shi, J. Zhao, and X. Xu, “A novel ensemble learning framework for stock forecasting with walk-forward validation,” Applied Intelligence, vol. 52, no. 5, pp. 4560–4578, 2022. doi: 10.1007/s10489-021-02579-5.
[23]R. Pardo, The Evaluation and Optimization of Trading Strategies, 2nd ed., Hoboken, NJ, USA: Wiley, 2008.