Volume 3 • Issue 2 • PP: 26–31 • 2026
A Dual-Bank Hybrid Predictive Model (DBHPM) for Financial Forecasting
Open Access & Copyright
© 2026 The Author(s). Published by ASPG. This article is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
Abstract
Forecasting financial performance is important for strategy formulation and identifying potential problems in banking institutions. This paper presents the Dual-Bank Hybrid Predictive Model (DBHPM), a financial forecasting model combining Multiple Linear Regression (MLR) and Random Forest Regression (RFR). It is validated on two actual financial datasets from Agrobank and NBU Bank covering 2021–2025. The model analyzes financial indicators including net profit, equity, and solvency, and forecasts them through 2027. Specifically, DBHPM uses MLR for linear modeling in the first step and RFR for nonlinear residuals in the second step; the paper reports stronger generalization and predictive strength than using the latter stage alone. Experiments show that DBHPM minimizes MAE and RMSE and achieves a coefficient of determination (R2) of 0.95 or above compared with models trained independently. The forecasts indicate net profit of approximately 1.18 billion sum for Agrobank and 3.66 billion sum for NBU Bank by the end of 2027. The proposed hybrid model supports predictive financial analysis for decision-making, risk awareness, and economic forecasting in the banking industry.
Keywords
References
[1] A. M. Ozbayoglu, M. U. Gudelek, and O. B. Sezer, “Deep learning for financial applications: A survey,” Applied Soft Computing, vol. 93, p. 106384, 2020.
[2] H. Wasserbacher and M. Spindler, “Machine learning for financial forecasting, planning and analysis: Re-cent developments and pitfalls,” Digital Finance, vol. 3, no. 1, pp. 1–21, 2021.
[3] E. Andreou, E. Ghysels, and A. Kourtellos, “Regression models with mixed sampling frequencies,” Journal of Econometrics, vol. 158, no. 2, pp. 246–261, 2010.
[4] E. Ghysels, P. Santa-Clara, and R. Valkanov, “Predicting volatility: How to get most out of returns data sampled at different frequencies,” Journal of Econometrics, vol. 131, no. 1–2, pp. 59–95, 2006.
[5] J. Bai, E. Ghysels, and J. Wright, “State space models and MIDAS regressions,” Econometric Reviews, vol. 32, no. 7, pp. 779–813, 2013.
[6] J. Chen, T. Chen, M. Shen, Y. Shi, and D. Wang, “Gated three-tower transformer for text-driven stock market prediction,” Multimedia Tools and Applications, vol. 81, no. 25, pp. 35 889–35 912, 2022.
[7] L. Yang, J. Li, R. Dong, Y. Zhang, and B. Smyth, “Deep learning in finance: A survey of applications and tech-niques,” Journal of Financial Data Science, vol. 4, no. 2, pp. 1–20, 2022.
[8] Linyi Yang, Jiazheng Li, Ruihai Dong, Yue Zhang, and Barry Smyth, “NumHTML: Numeric-oriented hierarchi-cal transformer model for multi-task financial forecast-ing,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 10, pp. 11 604–11 612, 2022.
[9] N. Nazareth, Y. V. Subrahmanyam, and S. S. Reddy, “Financial applications of machine learning: A literature review,” Expert Systems with Applications, vol. 219, p. 119640, 2023.
[10] Y. Li, S. Wang, H. Ding, and H. Chen, “Large languagemodels in finance: A survey,” 2023.
Cite This Article
Choose your preferred format
Publisher's Note
The statements, opinions, and data presented in this article are solely those of the author(s) and do not necessarily represent those of ASPG, the journal, or its editors. ASPG and the editors disclaim responsibility for any harm arising from the use of any ideas, methods, instructions, or products described in this article, to the fullest extent permitted by applicable law.