EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning and Machine Learning for Intelligent Stock Market Forecasting: A Management-Oriented Framework for Financial Decision-Making

Dr. Alok Kumar, Dr Aditi saxena, Dr.vinay raj R, Dr. K.P. Jabir Moosa, Dr Alikutty T P, Dr.Poonam Gupta

Computer and Decision Making: An International Journal · 04.11.2026 · 3 (6S) · 2719-2728

DOI: 10.1922/comdem.v3i6s.19012 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer and Decision Making: An International Journal, online erschienen am 04.11.2026. Verfasst von Dr. Alok Kumar und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The relevance of intelligent stock market forecasting to financial management has grown in significance, but this does not imply that the more complex the model, the better its out-of-sample performance. Decision support methods that are both accurate and informative about direction are needed for reliable forecasting. In this study, the conventional, machine learning, and deep learning methods of next-day NIFTY 50 forecasting were compared, and the appropriateness of the methods for making appropriate financial decisions for the management was evaluated. Data from the NIFTY 50 index for the daily time frame were used for the above period from January 2015 to April 2025, with the addition of Daily Return, SMA 10, EMA 10 and RSI 14 indicators. A train-test split of chronological 80:20 was used. The performance was compared against the benchmark of Naive Persistence by assessing the model using the metrics MAE, RMSE, MAPE, R-squared, and Directional Accuracy in the cases of Linear Regression, Random Forest, XGBoost and LSTM. Linear Regression recorded the lowest MAE of 125.44 and MAPE of 0.566%, with an RMSE of 180.82 and the highest directional accuracy of 53.74%. LSTM achieved an RMSE of 2278.71, while Random Forest and XGBoost achieved an RMSE of 4034.38 and 4049.66, respectively. The utility of Linear Regression thus had the best balance of price accuracy, direction, and interpretability, and was a testament to the fact that out-of-sample robustness may be a more significant metric than complexity for financial decision support.

Zitieren

APA

Kumar, D. A., saxena, D. A., R, D. R., Moosa, D. K. J., P, D. A. T., & Gupta, D. (2026). Deep Learning and Machine Learning for Intelligent Stock Market Forecasting: A Management-Oriented Framework for Financial Decision-Making. Computer and Decision Making: An International Journal, 3(6S), 2719-2728. https://doi.org/10.1922/comdem.v3i6s.19012

Vancouver

Kumar DA, saxena DA, R DR, Moosa DKJ, P DAT, Gupta D. Deep Learning and Machine Learning for Intelligent Stock Market Forecasting: A Management-Oriented Framework for Financial Decision-Making. Computer and Decision Making: An International Journal. 2026;3(6S):2719-2728. doi:10.1922/comdem.v3i6s.19012

BibTeX
@article{Kumar2026,
  author = {Kumar, Dr. Alok and saxena, Dr Aditi and R, Dr.vinay raj and Moosa, Dr. K.P. Jabir and P, Dr Alikutty T and Gupta, Dr.Poonam},
  title = {Deep Learning and Machine Learning for Intelligent Stock Market Forecasting: A Management-Oriented Framework for Financial Decision-Making},
  journal = {Computer and Decision Making: An International Journal},
  year = {2026},
  volume = {3},
  number = {6S},
  pages = {2719--2728},
  doi = {10.1922/comdem.v3i6s.19012}
}
Kostenlos starten →