Modification of the Random Forest Algorithm for Selecting Investment Instruments
https://doi.org/10.21822/2073-6185-2025-52-3-38-48
Abstract
Objective. The aim of the study is to improve the efficiency of investment decisions by developing a machine learning algorithm "Modified Random Forest" for forming an investment portfolio.
Method. The article proposes a binary classification algorithm "Modified Random Forest" based on the machine learning algorithms "Decision Tree" and "Random Forest". At the first stage, the algorithm builds a decision tree based on forecasts using the naive method and the ARIMA method, and at the second stage, it forms a "forest" of trees from random subsamples.
Result. The algorithm was tested on different time intervals on the instruments of the Russian Moscow Exchange index and the American NASDAQ index. To assess the accuracy of the algorithm, the Precision, Recall, and Accuracy metrics were selected. For comparison, shares were selected for the portfolio under the same conditions using the classical Markowitz problem method, the results of which for the corresponding metrics were somewhat weaker. At the same time, the proposed algorithm takes significantly less time to run, since it does not solve the optimization problem when forming a portfolio.
Conclusion. The developed machine learning algorithm "Modified random forest", based on the Arima forecasting methods and naive forecast allows solving the problem of increasing the efficiency of investment portfolio management, and due to its binary nature, it can be used not only in the financial sector, but also for the task of classifying any other predicted objects.
About the Author
A. V. Zinenko
Siberian Federal University
Russian Federation
Russian Federation
Anna V. Zinenko - Cand. Sci. (Eng.), Assoc. Prof.; Department of Economic and Financial Security.
79 Svobodny Ave., Krasnoyarsk 660041
References
1. Bodie, Z., Kane, A., Marcus, A. Investment Principles: trans. from English. Moscow: Williams Publishing House. 2005. 984 p. ISBN 9785845913111
2. Kasimov, Yu.F. Financial Investments. Moscow: Ankil. 2008. 230 p.
3. Ivanchenko, I.S., Osei, D.D. Optimizing the Structure of Russian Gold and Foreign Exchange Reserves Using the Black-Litterman Model. Financial Journal. 2018; 1 (41): 26–38. DOI: 10.31107/2075-1990-2018-1-26-38 (In Russ)
4. Sharpe, W., Bailey, A. Investments. Trans. from English. Moscow: INFRA-M. 2009. 1027 p. ISBN 9785160167893 (In Russ)
5. Kosorukov O.A., Mishchenko A.V. Operations Research. Moscow: Examen. 2003. 448 p.
6. Zinenko A.V., Martynova T.A. Modified knapsack method algorithm for solving the inverse Markowitz problem. Financial Economics. 2024; 6: 136 –139 (In Russ)
7. Box J., Jenkins G. Time Series Analysis: Forecast and Management. Moscow: Mir. 1974. 820 p. ISBN 2100013644333 (In Russ)
8. Vyugin V.V. Mathematical Foundations of Machine Learning and Forecasting. Moscow: MCNO. 2024. 484 p. ISBN 9785443920146 (In Russ)
9. Zinenko A., Stupina A. Financial time series forecasting methods. ITM Web Conf. II International Workshop “Hybrid Methods of Modeling and Optimization in Complex Systems” (HMMOCS-II 2023). 2024; 59: 02005. DOI: 10.1051/itmconf/20245902005
10. Chou J.-Sh., Chen K.-E. Optimizing investment portfolios with a sequential ensemble of decision tree-based models and the FBI algorithm for efficient financial analysis. Applied Soft Computing. 2024; 158: 111550. DOI: 10.1016/j.asoc.2024.111550
11. Marchenko A.L. Python: a big book of examples. Moscow: Moscow University Press. 2023;361 (In Russ)
12. Brink H., Richards D., Feverloff M. Machine Learning. St. Petersburg: Peter Ilib. 2017. 336 p. ISBN 9785496029896 (In Russ)
13. Rashka S., Mirjalili R. Python and Machine Learning. Moscow: Packt. 2020. 656 p. . ISBN 9785970604090 (In Russ)
14. Carter D. Artificial Intelligence. Machine Learning. Moscow: "Author". 2024. 268 p. (In Russ)
15. Habbab F., Kampouridis M. Anin-depth investigation of five machine learning algorithms for optimizing mixed-asset portfolios including REITs. Systems With Applications. 2024; 235:121102. DOI: 10.1016/j.eswa.2023.121102
16. Jun S.Y. et al. Stock investment strategy combining earnings power index and machine learning. International Journal of Accounting Information Systems. 2022; 47: 100576. DOI: 10.1016/j.accinf.2022.100576
17. Behera J. et al. Prediction based mean-value-at-risk portfolio optimization using machine learning regression algorithms for multi-national stock markets. Engineering Applications of Artificial Intelligence. 2023; 120:105843. DOI: 10.1016/j.engappai.2023.105843
18. 190. Parisi L., Manaog M.L. Optimal Machine Learningand Deep Learning-driven algorithms for predicting the future value of investments: A systematic review and meta-analysis// Engineering Applications of Artificial Intelligence. 2025; 142: 109924. DOI: 10.1016/j.engappai.2024.109924
19. Ramirez H.E., Serrano R. Optimal investment with insurable background risk and nonlinear portfolio allocation frictions. Applied Mathematics and Computation. 2025; 485: 129023. DOI: 10.1016/j.amc.2024.129023
20. Lin W., Taamouty A. Portfolio selection under non-gaussianity and systemic risk: A machine learning based forecasting approach. International Journal of Forecasting. 2024; 40: 1179 – 1188. DOI: 10.1016/j.ijforecast.2023.10.007
Review
For citations:
Zinenko A.V. Modification of the Random Forest Algorithm for Selecting Investment Instruments. Herald of Dagestan State Technical University. Technical Sciences. 2025;52(3):38-48. (In Russ.) https://doi.org/10.21822/2073-6185-2025-52-3-38-48
Views:
6
Email this article 
