Improving the Markowitz Efficient Boundary Using Random Forest -an analytical study in Iraq stock exchange
الملخص
The current research aims to determine the efficient Markowitz boundary shape after applying artificial intelligence techniques. The problem identified is the randomness of statistical distributions following non-linear movements, which hinders the application of linear statistical models. This necessitated the use of random forest techniques. A review of previous studies revealed their shortcomings in not employing random forest techniques to improve the efficient boundary shape within the Iraqi context, The study population consisted of all companies listed in the Iraq Stock Exchange from 2020 to 2024, while the sample was selected from companies that consists 42 companies meet the condition of continuity of listing and the loss rate does not exceed 20% , with 42 companies that fully met the listing requirements However, the problem of missing values still remains, and it was addressed by using a mask and padding tool to handle the missing data, we use the traditional financial method for building efficient frontier once and two by Random forest, The PyCharm software was used to analyze the data and extract the results, the main result are the ability of random forests to overcome the problem of randomness and large fluctuations in values, explore non-linear relationships ,significantly improving the efficient limit, especially by pruning the efficient limit to touch the capital market line with a larger sum of points
التنزيلات
المراجع
[1] M. Kasimovas, “Integrating Machine Learning and Portfolio Optimization Methods for Enhancing Baltic Stock Market Portfolio Composition Analysis Masters’s Final Degree Project,” 2024.
[2] N. Alexander and W. Scherer, “Using Machine Learning to Forecast Market Direction with Efficient Frontier Coefficients,” J. Financ. Data Sci., vol. 5, no. 3, pp. 9–22, 2023, doi: 10.3905/jfds.2023.1.128.
[3] J. S. Masuda, “Portfolio Optimization Using a Hybrid Machine Learning Stock Selection Model,” 2024.
[4] E. Grass, J. Ortmann, B. Balcik, and W. Rei, “A machine learning approach to deal with ambiguity in the humanitarian decision-making,” Prod. Oper. Manag., vol. 32, no. 9, pp. 2956–2974, 2023, doi: 10.1111/poms.14018.
[5] M. Pal et al., “Revolutionizing Active Investing With Machine Learning,” SSRN Electron. J., 2024, doi: 10.2139/ssrn.4675439.
[6] H. Zhao, Predicting Stock Prices and Optimizing Portfolios: A Random Forest and Monte Carlo-Based Approach Using NASDAQ-100, no. Icdeba 2024. Atlantis Press International BV, 2025. doi: 10.2991/978-94-6463-652-9_95.
[7] V. Rathi, M. Kshirsagar, and C. Ryan, “Enhancing Portfolio Performance: A Random Forest Approach to Volatility Prediction and Optimization,” Int. Conf. Agents Artif. Intell., vol. 3, no. Icaart, pp. 1278–1285, 2024, doi: 10.5220/0012464600003636.
[8] E. Moyoweshumba and M. Seitshiro, “Leveraging Markowitz, random forest, and XGBoost for optimal diversification of South African stock portfolios,” Data Sci. Financ. Econ., vol. 5, no. 2, pp. 205–233, 2025, doi: 10.3934/DSFE.2025010.
[9] H. Zhang, “Limitations and Critique of Modern Portfolio Theory: A Comprehensive Literature Review,” Adv. Econ. Manag. Polit. Sci., vol. 60, no. 1, pp. 24–29, Jan. 2024, doi: 10.54254/2754-1169/60/20231148.
[10] A. Fernández and S. Gómez, “Portfolio selection using neural networks,” Comput. Oper. Res., vol. 34, no. 4, pp. 1177–1191, 2007, doi: 10.1016/j.cor.2005.06.017.
[11] H. Tan, “An Empirical Study on the Markowitz portfolio,” BCP Bus. Manag., vol. 44, pp. 503–511, 2023, doi: 10.54691/bcpbm.v44i.4861.
[12] K. Doerner, C. Stummer, and C. Strauss, “Ant Colony Optimization in Multiobjective Portfolio Selection,” Proc. 4th Metaheuristics Int. Conf., no. January, pp. 243–248, 2001.
[13] G. F. Deng and W. T. Lin, “Ant colony optimization for Markowitz mean-variance portfolio model,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 6466 LNCS, pp. 238–245, 2010, doi: 10.1007/978-3-642-17563-3_29.
[14] M. Li and Y. Wu, “Dynamic Decision Model of Real Estate Investment Portfolio Based on Wireless Network Communication and Ant Colony Algorithm,” Wirel. Commun. Mob. Comput., vol. 2021, 2021, doi: 10.1155/2021/9261312.
[15] L. O. Bettencourt, A. Petukhina, and A. Tetereva, “Advancing Markowitz: Asset Allocation Forest.” [Online]. Available: https://ssrn.com/abstract=4781685
[16] J. Zheng, D. Xin, Q. Cheng, M. Tian, and L. Yang, “The Random Forest Model for analyzing and Forecasting the US Stock Market under the background of smart finance,” 2024, pp. 82–90. doi: 10.2991/978-94-6463-419-8_11.
[17] E. Kouloumpris and I. Vlahavas, “Markowitz random forest: Weighting classification and regression trees with modern portfolio theory,” Neurocomputing, vol. 620, no. December 2024, 2025, doi: 10.1016/j.neucom.2024.129191.
[18] M. Rahiminezhad Galankashi, F. Mokhatab Rafiei, and M. Ghezelbash, Portfolio selection: a fuzzy-ANP approach, vol. 6, no. 1. Financial Innovation, 2020. doi: 10.1186/s40854-020-00175-4.
[19] E. Moyoweshumba and M. Seitshiro, “Leveraging Markowitz, random forest, and XGBoost for optimal diversification of South African stock portfolios,” Data Sci. Financ. Econ., vol. 5, no. 2, pp. 205–233, 2025, doi: 10.3934/dsfe.2025010.
[20] M. G. † G. P. ‡ M. Pedio, “Chapter 1: Machine Learning in Portfolio Decisions,” Artif. Intell. beyond Financ., vol. 15, no. October, pp. 1–72, 2024.
[21] B. Nandi, S. Jana, and K. P. Das, “Machine learning-based approaches for financial market prediction: A comprehensive review,” J. AppliedMath, vol. 1, no. 2.1, p. 134, Aug. 2023, doi: 10.59400/jam.v1i2.134.
[22] A. Chaweewanchon and R. Chaysiri, “Markowitz Mean-Variance Portfolio Optimization with Predictive Stock Selection Using Machine Learning,” Int. J. Financ. Stud., vol. 10, no. 3, Sep. 2022, doi: 10.3390/ijfs10030064.
[23] P. Humairah and D. Agustina, “Stock Price Index Prediction Using Random Forest Algorithm for Optimal Portfolio,” J. Varian, vol. 8, no. 1, pp. 113–124, 2024, doi: 10.30812/varian.v8i1.4276.
[24] X. Lyu, “Portfolio Optimization Strategies: New Approaches Based on Machine Learning Forecasting,” Highlights Business, Econ. Manag., vol. 40, pp. 1077–1082, 2024, doi: 10.54097/cpgtg807.
[25] and J. F. T. Hastie, R. Tibshirani, “The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd ed. New York, NY, USA: Springer, 2009.,” Math. Intell., vol. 27, no. 2, pp. 83–85, 2009.
