Evaluating the Predictive Power of Deep Learning Models in Financial Markets
DOI:
https://doi.org/10.58575/nkpn1t16Keywords:
Deep Learning Models , Stock price, prediction, CNN, LSTM, GRU, ForecastingAbstract
Stock price forecasting remains a prominent area of research in financial markets due to the complicated, nonlinear, and volatile nature of stock price paths. While high-performance deep learning models like Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRU) demonstrate superiority in financial forecasting, their comparative efficacy in emerging markets, such as the KSE-100 Index, remains inadequately explored. This study examines the prediction performance of these models based on past stock prices of the KSE-100 Index and based on standard performance measures (root mean square error, mean absolute error, and mean absolute percentage error) to provide a holistic assessment of the models. Our research indicates that GRU consistently performs better than CNN and LSTM, with lower errors and higher predictive accuracy. This work enhances current forecasting methods by utilizing advanced data preparation, feature engineering, and hyperparameter optimization to augment model performance. The results have practical implications for policymakers and investors in emerging markets by emphasizing the applicability of deep learning to financial decision-making.
Downloads
References
Cho, K., van Merri¨ enboer, B., Bahdanau, D., & Bengio, Y. (2014). On the properties of neural machine translation: Encoder–decoder approaches. Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation, 103–111. https://doi.org/10.3115/v1/W14-4012
Deng, H. (2024). The advancements and progresses of artificial intelligence-based keyword extraction methods. Proceedings of the 2024 International Conference on Artificial Intelligence and Communication, 580–585. https://doi.org/10.2991/978-94-6463-512-6 61
Gao, Y., Wang, R., & Zhou, E. (2021). Stock prediction based on optimized LSTM and GRU models. Scientific Programming, 2021, 1–8. https://doi.org/10.1155/2021/5512678
Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
Hussain, B., Afzal, M. K., Ahmad, S., & Mostafa, A. M. (2021). Intelligent traffic flow prediction using optimized GRU model. IEEE Access, 9, 100736–100746. https://doi.org/10.1109/ACCESS.2021.3096597
Jia, P., Liu, H., Wang, S., & Wang, P. (2020). Research on a mine gas concentration forecasting model based on a GRU network. IEEE Access, 8, 38023–38031. https://doi.org/10.1109/ACCESS.2020.2975601
Kexin, H., & Zhijin, Z. (2022). Prediction stock price based on CNN and LSTM models. Financial Engineering and Risk Management, 5(7), 14–21. https://www.scirp.org/journal/paperinformation.aspx?paperid=116988
Kim, H. Y., & Won, C. H. (2018). Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models. Expert Systems with Applications, 103, 25–37. https://doi.org/10.1016/j.eswa.2018.03.008
Ku, C. S., Xiong, J., Chen, Y.-L., Cheah, S. D., Soong, H. C., & Por, L. Y. (2023). Improving stock market predictions: An equity forecasting scanner using long short-term memory method with dynamic indicators for Malaysia stock market. Mathematics, 11(11), 2470. https://doi.org/10.3390/math11112470
LeCun, Y., & Bengio, Y. (1995). Convolutional networks for images, speech, and time series. In M. A. Arbib (Ed.), The handbook of brain theory and neural networks (pp. 255–258). MIT Press. https://doi.org/10.5555/303568.303704
Noorbakhsh, A., & Shaygani, M. (2024). Deep learning approach for stock price prediction: Comparing single and hybrid models. Advances in Industrial Engineering, 58(1), 237–249. https://doi.org/10.22059/aie.2024.372470.1889
Pokhrel, N. R., Dahal, K. R., Rimal, R., Bhandari, H. N., Khatri, R. K., Rimal, B., & Hahn, W. E. (2022). Predicting NEPSE index price using deep learning models. Machine Learning with Applications, 9, 100385. https://doi.org/10.1016/j.mlwa.2022.100385
Rita, N. O. (2019). Machine learning techniques for predicting the stock market using daily market variables. https://run.unl.pt/bitstream/10362/94992/1/TGI0294.pdf
Roondiwala, M., Patel, H., & Varma, S. (2017). Predicting stock prices using LSTM. International Journal of Science and Research, 6(4), 1754–1756. https://www.ijsr.net/archive/v6i4/ART20174612.pdf
Xiuyun, G., Ying, W., Yang, G., Chengzhi, S., Wen, X., & Yimiao, Y. (2018). Short-term load forecasting model of gru network based on deep learning framework. 2018 2nd IEEE Conference on Energy Internet and Energy System Integration. https://doi.org/10.1109/EI2.2018.8602652
Zafar, N., Haq, I. U., Chughtai, J.-u.-R., & Shafiq, O. (2022). Applying hybrid LSTM-GRU model based on heterogeneous data sources for traffic speed prediction in urban areas. Sensors, 22(9), 3348. https://doi.org/10.3390/s22093348
Zaheer, S., Anjum, N., Hussain, S., Algarni, A. D., Iqbal, J., Bourouis, S., & Ullah, S. S. (2023). A multi parameter forecasting for stock time series data using LSTM and deep learning model. Mathematics, 11(3), 590. https://doi.org/10.3390/math11030590
