Evaluation of Accounting Criteria in Choosing the Optimal Stock Portfolio in Iran's Capital Market: A Hybrid SVM-DEA Approach
Keywords:
Iranian Capital Market, Stock Portfolio Selection, Accounting Indicators, Support Vector Machine (SVM), Data Envelopment Analysis (DEA), Firm EfficiencyAbstract
The present study aimed to evaluate the efficiency of accounting indicators in selecting an optimal stock portfolio and to develop a hybrid Support Vector Machine (SVM) and Data Envelopment Analysis (DEA) framework for improving investment decision-making in Iran’s capital market. This applied quantitative study was conducted using data from firms listed on the Tehran Stock Exchange during the period 2009–2022. The sample was selected through systematic elimination criteria. In the first stage, accounting variables extracted from corporate financial statements were screened using the Support Vector Machine algorithm to identify and rank the most influential indicators affecting stock returns. In the second stage, the selected accounting indicators together with risk variables were incorporated into an output-oriented BCC Data Envelopment Analysis model with variable returns to scale to evaluate the relative efficiency of firms. Finally, an optimal stock portfolio was constructed based on efficiency scores and its performance was assessed. The results revealed that the price-to-book ratio (P/B) was the most influential accounting indicator with a weight of 0.639, followed by sales growth (0.569) and earnings growth (0.541). The SVM model demonstrated satisfactory classification performance in distinguishing efficient from inefficient firms (AUC = 0.81). DEA results indicated that approximately 23% of firms were located on the efficiency frontier, whereas 77% were classified as inefficient. Efficiency gap analysis showed that most inefficiencies were associated with weaknesses in liquidity indicators and capital productivity. Furthermore, market-based indicators, particularly P/B and E/P ratios, played the most significant role in determining firm efficiency and stock selection. The findings suggest that the hybrid SVM-DEA model outperforms traditional approaches in screening relevant accounting variables, identifying efficient firms, and constructing optimal stock portfolios. Integrating machine learning techniques with efficiency evaluation models can enhance investment decision accuracy and improve portfolio performance in the Iranian capital market.
Downloads
References
Al Janabi, M. A. (2021). Multivariate Portfolio Optimization under Illiquid Market Prospects: A Review of Theoretical Algorithms and Practical Techniques for Liquidity Risk Management. Journal of Modelling in Management, 16(1), 288-309. https://doi.org/10.1108/JM2-07-2019-0178
Anuno, F., Madaleno, M., & Vieira, E. (2024). Testing of Portfolio Optimization by Timor-Leste Portfolio Investment Strategy on the Stock Market. Journal of Risk and Financial Management, 17(2), 78. https://doi.org/10.3390/jrfm17020078
Behera, J., Pasayat, A. K., Behera, H., & Kumar, P. (2023). Prediction based mean-value-at-risk portfolio optimization using machine learning regression algorithms for multi-national stock markets. Engineering Applications of Artificial Intelligence, 120, 105843. https://doi.org/10.1016/j.engappai.2023.105843
Dunis, C. L., Middleton, P. W., Karathanasopolous, A., & Theofilatos, K. (2019). Artificial Intelligence in Financial Markets. Springer. http://repo.darmajaya.ac.id/5259/1/Artificial%20Intelligence%20in%20Financial%20Markets_%20Cutting%20Edge%20Applications%20for%20Risk%20Management%2C%20Portfolio%20Optimization%20and%20Economics%20%28%20PDFDrive%20%29.pdf
Jaiyeoba, H. B., Abdullah, M. A., & Ibrahim, K. (2020). Optimizing Stock Portfolio Performance with a Combined RG1-TOPSIS Model: Insights from the Chinese Market. Journal of the Knowledge Economy, 11(4), 1600-1622. https://doi.org/10.1007/s13132-019-00614-4
Khalouzadeh, H., & Amiri, N. (2006). Optimal Stock Portfolio Selection in the Iranian Stock Market Based on Value-at-Risk Theory. Economic Research(73), 211-231. https://www.magiran.com/paper/361447/optimal-portfolio-selection-in-iran-stock-exchange-via-value-at-risk-theory?lang=en
Kim, H. Y., & Ahn, H. (2021). Ant colony optimization for portfolio selection in volatile markets. Journal of Computational Finance, 24(1), 1-18.
Larni-Fooeik, A., Sadjadi, S. J., & Mohammadi, E. (2024). Stochastic portfolio optimization: A regret-based approach on volatility risk measures: An empirical evidence from The New York stock market. PLoS One, 19, e0299699. https://doi.org/10.1371/journal.pone.0299699
Leung, P. l., Ng lui, K., & wong, W. (2022). An improved estimation to make Markowitz's portfolio optimization theory users friendly and estimation accurate with application on the US stock market investment. 85-98. https://doi.org/10.1016/j.ejor.2012.04.003
Levchenko, V., & Ostapenko, M. (2016). Formation of the Optimal Portfolio of Insurer’s Services of the Voluntary Types of Insurance. Insurance Markets and Companies, 7(1), 45-51. https://doi.org/10.21511/imc.7(1).2016.05
Li, B., & Teo, K. L. (2021). Portfolio optimization in real financial markets with both uncertainty and randomness. Applied Mathematical Modelling, 100, 125-137. https://doi.org/10.1016/j.apm.2021.08.006
Mostafaei Darmian, S., & Doaei, M. (2021). A Stochastic Optimization-Based Approach for Solving the Portfolio Selection Problem in Iran's Capital Market Using Metaheuristic Algorithms. Quarterly Journal of Applied Economic Theories, 8(4).
Navidi, H. R., Nejoomi Markid, A., & Mirzazadeh, H. (2010). Portfolio Selection in Tehran Stock Exchange Market with a Genetic Algorithm. Journal of Economic Research (Tahghighat- E- Eghtesadi), 44(4). https://jte.ut.ac.ir/article_20348.html
Redkin, N. (2019). Investment Portfolio Optimization on Russian Stock Market in Context of Behavioral Theory. Finance Theory and Practice, 23(4), 99-116. https://doi.org/10.26794/2587-5671-2019-23-4-99-116
Salemi Najafabadi, M., Saadatfar, N., & Karimi, F. (2014). Forecasting returns on investment opportunities in Iran's financial markets considering market interactions and forming an optimal investment portfolio using artificial intelligence. Asset Management and Financing, 2(4), 35-50. https://doi.org/10.22108/amf.2014.19897
Wang, Z. (2025). Active vs. Passive Investment in the Post-Pandemic U.S. Stock Market: A Sharpe Ratio-Based Portfolio Optimization Compared to the S&P 500. Highlights in Business Economics and Management, 61, 41-45. https://doi.org/10.54097/j6kd1r14
Downloads
Published
Submitted
Revised
Accepted
Issue
Section
License
Copyright (c) 2025 Hossein Nabieiboroujeni (Author); Hamidreza Jafari Dehkordi (Corresponding author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
