Forecasting Stock Market Trends with LSTM-Based Deep Learning Models

References

[1] T. Bollerslev, "Generalized autoregressive conditional heteroskedasticity," Journal of Econometrics, vol. 31, no. 3, pp. 307-327, 1986.

[2] S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Computation, vol. 9, no. 8, pp. 1735-1780, 1997.

[3] T. Fischer and C. Krauss, "Deep learning with long short-term memory networks for financial market predictions," European Journal of Operational Research, vol. 270, no. 2, pp. 654-669, 2018.

[4] G. E. P. Box, G. M. Jenkins, G. C. Reinsel, and G. M. Ljung, Time Series Analysis: Forecasting and Control, 5th ed. Hoboken, NJ: Wiley, 2015.

[5] T. Chen and C. Guestrin, "XGBoost: A scalable tree boosting system," in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, San Francisco, CA, 2016, pp. 785-794.

[6] K. Cho et al., "Learning phrase representations using RNN encoder-decoder for statistical machine translation," in Proc. EMNLP, Doha, Qatar, 2014, pp. 1724-1734.

[7] 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.

[8] X. Ding, Y. Zhang, T. Liu, and J. Duan, "Deep learning for event-driven stock prediction," in Proc. 24th Int. Joint Conf. Artificial Intelligence, Buenos Aires, 2015, pp. 2327-2333.

[9] Y. Li, Y. Bu, J. Li, and H. Liu, "Multi-task learning for stock prediction with attention-based LSTM," in Proc. IEEE Int. Conf. Systems, Man, and Cybernetics, Bari, Italy, 2019, pp. 3650-3655.

[10] B. N. Lim, S. O. Arik, N. Loeff, and T. Pfister, "Temporal fusion transformers for interpretable multi-horizon time series forecasting," International Journal of Forecasting, vol. 37, no. 4, pp. 1748-1764, 2021.

[11] J. Kim and J. Kim, "Forecasting stock prices with a feature fusion LSTM-CNN model using different representations of the same data," PLOS ONE, vol. 14, no. 2, p. e0212320, 2019.

[12] A. Graves, A. Mohamed, and G. Hinton, "Speech recognition with deep recurrent neural networks," in Proc. IEEE ICASSP, Vancouver, BC, Canada, 2013, pp. 6645-6649.

[13] D. Shah, H. Isah, and F. Zulkernine, "Stock market analysis: A review and taxonomy of prediction techniques," International Journal of Financial Studies, vol. 7, no. 2, p. 26, 2019.

[14] A. Vaswani et al., "Attention is all you need," Advances in Neural Information Processing Systems, vol. 30, 2017.

[15] R. Adhikari and R. K. Agrawal, "An introductory study on time series modeling and forecasting," arXiv preprint arXiv:1302.6613, 2013.

[16] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, vol. 521, no. 7553, pp. 436-444, 2015.

[17] Y. Gal and Z. Ghahramani, "A theoretically grounded application of dropout in recurrent neural networks," Advances in Neural Information Processing Systems, vol. 29, 2016.

[18] W. Bao, J. Yue, and Y. Rao, "Forecasting stock market using deep learning and technical analysis," PLOS ONE, vol. 13, no. 4, 2018.

[19] Z. Hu et al., "Listening to chaotic whispers: A deep learning framework for news-oriented stock trend prediction," in Proc. 11th ACM WSDM, Marina Del Rey, CA, 2018, pp. 261-269.

[20] J. Yao, C. L. Tan, and H. Poh, "Neural networks for technical analysis: A study on KLCI," International Journal of Theoretical and Applied Finance, vol. 2, no. 2, pp. 221-241, 1999.

@article{1716441,
      author = {Prince Verma, Samriddhi Singh},
      title = {Forecasting Stock Market Trends with LSTM-Based Deep Learning Models},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {9},
      number = {10},
      pages = {1889-1896},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1716441.pdf},
      abstract = {Forecasting movements in the stock market has long been one of the most desired yet frustrating tasks in data science. Stock markets are volatile, emotional, and driven by factors that no individual model could ever hope to capture entirely. In this study, we analyze and compare how well LSTM neural networks can predict stock movement trends and, more specifically, what impact the architecture itself may have. Using ten years of daily trade data (from 2014 through 2024) for three stocks – Apple Inc. (AAPL), Reliance Industries (RELIANCE.NS), and NIFTY 50 index, we have compared how well five different architectures – starting with a simple one-layer LSTM to a more sophisticated approach of a three-layer stacked LSTM with attention mechanisms – fared in their predictions. Our feature selection ranges from technical analysis tools, such as RSI, MACD, and Bollinger bands, to economic indicators, like VIX and spread between treasury yields. Of our models, the stacked LSTM with attention turned out to have the best performance, with a MAPE of 1.73% and accuracy in predicting price direction of 84.6%, significantly better than any other model tried (ARIMA, GRU, XGBoost, LSTMs). We have tested our models on the stock market crash caused by COVID-19 in early 2020 to see their performance during out-of-domain situations, i.e., when market conditions drastically change, as well as what those results imply about the use of these models in practice.},
      keywords = {LSTM, Stock Market Prediction, Deep Learning, Time Series Forecasting, Technical Indicators, Attention Mechanism, Recurrent Neural Network, Financial Analytics},
      month = {April},
      doi = {https://doi.org/10.64388/IREV9I10-1716441}
  }