Authors : Sujit Shibaprasad Maity

Volume/Issue : Volume 11 - 2026, Issue 3 - March

Google Scholar : https://tinyurl.com/s93494vr

Scribd : https://tinyurl.com/43fsyj3m

DOI : https://doi.org/10.38124/ijisrt/26mar1068

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Accurate and timely stock market analysis is essential for informed investment decision-making; however, market volatility, noise, and non-stationary behavior make forecasting a challenging problem. This paper presents StockBuddy Assistant, a web-based decision support system that integrates real-time stock data acquisition, exploratory visualization, and short-term price forecasting within an interactive user interface. The system retrieves historical daily stock data via the Alpha Vantage API and applies a supervised machine learning approach using Linear Regression to forecast closing prices for the next seven trading days. The forecasting pipeline explicitly accounts for trading-day constraints by excluding weekends. In addition to visualization, a rule-based recommendation module provides basic investment guidance based on forecasted trends. The proposed system functions as an interpretable expert decision support system, emphasizing deployment efficiency and real-time usability for practical financial analytics. Experimental observations across multiple equities demonstrate that while linear models can capture short-term directional trends under stable conditions, they remain limited in highly volatile markets. The proposed system emphasizes interpretability, accessibility, and extensibility, serving as a foundation for more advanced predictive and risk-aware financial analytics. A rolling-window backtesting framework is employed to evaluate forecasting robustness, and performance is benchmarked against a naïve persistence baseline. Quantitative results demonstrate improved mean absolute error and directional accuracy while maintaining sub-millisecond inference latency suitable for real-time deployment.

Keywords : Stock Market Forecasting, Financial Time Series, Linear Regression, Decision Support Systems, Streamlit, Fintech Analytics.

References :

1. R. H. Shumway and D. S. Stoffer, Time Series Analysis and Its Applications: With R Examples, 4th ed. New York, NY, USA: Springer, 2017.
2. T. Bollerslev, “Generalized autoregressive conditional heteroskedasticity,” Journal of Econometrics, vol. 31, no. 3, pp. 307–327, 1986.
3. I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA, USA: MIT Press, 2016.
4. Alpha Vantage, “Stock Market APIs.” [Online]. Available: https://www.alphavantage.co/
5. T. Singh, S. Choudhary, and P. Kumar, “An efficient real-time stock prediction exploiting incremental and offline–online learning,” Scientific Reports, vol. 12, no. 1, 2022.
6. S.-J. Yu, “The Design of an Intelligent Lightweight Stock Trading System,” Applied Sciences, vol. 13, no. 4, 2023.
7. D. Campos, M. W. R. da Silva, and J. Gama, “LightTS: Lightweight Time Series Classification with Knowledge Distillation,” in Proceedings of the ACM Conference on Information and Knowledge Management (CIKM), 2023.
8. H. Subramanian, A. Gupta, and R. Bansal, “A decision support system using signals from social media for stock market analysis,” Decision Support Systems, vol. 173, 2024.
9. G. Kostopoulos, E. P. Kastritis, and P. Pintelas, “Explainable Artificial Intelligence-Based Decision Support Systems: A Survey,” Electronics, vol. 13, no. 2, 2024.