Modeling the Dynamics of Financial Markets Using Artificial Intelligence Approaches
Keywords:
Financial Market Dynamics, Artificial Intelligence, LSTM, DCC, GARCH, Volatility, Intermarket Dependence, Tehran Stock ExchangeAbstract
This study aimed to model the dynamics and interdependence of the Tehran Stock Exchange, Iran Fara Bourse, Iran Mercantile Exchange, and Iran Energy Exchange and to evaluate the predictive capacity of artificial intelligence-based LSTM models for forecasting financial returns. This applied quantitative study was based on financial time-series analysis. Weekly data from 2015 to 2025 were collected for four return series, including the Tehran Stock Exchange Overall Index, Iran Fara Bourse Index, Iran Mercantile Exchange Company stock, and Iran Energy Exchange Company stock, yielding 492 synchronized observations per series. Logarithmic returns were calculated after data preprocessing and calendar synchronization. Stationarity was assessed using the Augmented Dickey–Fuller test. Volatility regimes, directional intermarket relationships, impulse response functions, and time-varying dependencies were examined using regime-based analysis, causality tests, and DCC-GARCH modeling. Univariate and multivariate long short-term memory models were then developed. The first 70% of observations were used for training and the final 30% for out-of-sample testing. Forecast performance was evaluated using RMSE, MAE, MAPE, and R². Significant bidirectional causality was observed between the Tehran Stock Exchange and Iran Fara Bourse, while the Tehran Stock Exchange also significantly influenced the mercantile and energy exchange proxies. Impulse response analysis showed that shocks originating in the Tehran Stock Exchange generated the strongest responses in other markets. DCC-GARCH results indicated persistent and time-varying conditional correlations, with the strongest dependence between the Tehran Stock Exchange and Iran Fara Bourse. Multivariate LSTM models consistently outperformed univariate models across all four markets, producing lower RMSE, MAE, and MAPE values and higher R² values. The Iranian financial market exhibits dynamic interdependence, asymmetric shock transmission, and time-varying correlations, and incorporating cross-market information into multivariate LSTM models improves forecasting performance compared with models based only on each market’s own historical returns.
References
[1] G. C. Mara, Y. Kumar, V. P. K, S. Madan, and R. A. M. Chandana, "Advance AI and Machine Learning Approaches for Financial Market Prediction and Risk Management: A Comprehensive Review," Journal of Computer Science and Technology Studies, vol. 7, no. 4, pp. 727-749, 2025, doi: 10.32996/jcsts.2025.7.4.86.
[2] G. K. Shukla, "Revolutionizing Trading: Unlocking the Potential of Artificial Intelligence in Financial Markets"," International Journal of Engineering Applied Sciences and Technology, vol. 09, no. 02, pp. 111-114, 2024, doi: 10.33564/ijeast.2024.v09i02.010.
[3] B. Odonkor, S. Kaggwa, P. U. Uwaoma, A. O. Hassan, and O. A. Farayola, "Integrating Artificial Intelligence in Accounting: A Quantitative Economic Perspective for the Future of US Financial Markets," Finance & Accounting Research Journal, vol. 6, no. 1, pp. 56-78, 2024, doi: 10.51594/farj.v6i1.723.
[4] F. Khan, S. Ullah Jan, and H. M. Zia-ul-Haq, "Artificial Intelligence Adoption, Audit Quality and Integrated Financial Reporting in GCC Markets," Asian Review of Accounting, vol. 33, no. 3, pp. 464-495, 2025, doi: 10.1108/ARA-03-2024-0085.
[5] A. Zeinali Kermani, H. Piri, A. Payan, and R. Sotudeh, "Investigating the Role of Blockchain Technology and Artificial Intelligence in the Enhancement and Improvement of Accounting Information Systems Using an Interpretive Structural Approach," Business, Marketing, and Finance Open, vol. 2, no. 4, pp. 1-17, 2025, doi: 10.61838/bmfopen.2.4.8.
[6] A. M. Aghadoust and M. Behzadnia, "Measuring the Impact of Artificial Intelligence on Financial Performance Considering the Mediating Effect of Sensory Marketing Effectiveness," 2025.
[7] J. Mansilla-Lopez, "Factors, Forecasts, and Simulations of Volatility in the Stock Market Using Machine Learning," Journal of Risk and Financial Management, 2025, doi: 10.3390/jrfm18050227.
[8] H. Md Arman, "Artificial Intelligence-Based Financial Analysis Models for Predicting Market Risk and Investment Decisions in US Companies," ASRC. Procedia: Global Perspectives in Science and Scholarship, vol. 1, no. 01, pp. 1066-1095, 2025, doi: 10.63125/9csehp36.
[9] F. Ghallabi, B. Souissi, A. M. Du, and S. Ali, "ESG stock markets and clean energy prices prediction: Insights from advanced machine learning," International Review of Financial Analysis, vol. 97, no. 1, p. 103889, 2025, doi: 10.1016/j.irfa.2024.103889.
[10] U. O. Ogbuonyalu, K. Abiodun, S. Dzamefe, E. N. Vera, A. Oyinlola, and I. Emmanuel, "Assessing Artificial Intelligence Driven Algorithmic Trading Implications on Market Liquidity Risk and Financial Systemic Vulnerabilities," pp. 18-21, 2024, doi: 10.38124/ijsrmt.v3i4.433.
[11] M. H. Poostforoush, A. Monajemi, S. Daei-Karimzadeh, and S. Samadi, "Automation of Algorithmic Trading Strategies in Artificial Financial Markets by Combining Machine Learning Techniques and Agent-based Modeling," Iranian Journal of Finance, vol. 9, no. 3, pp. 95-134, 2025, doi: 10.30699/ijf.2025.458379.1472.
[12] H. Xia, S. Sun, X. Wang, and B. An, "Market-Gan: Adding Control to Financial Market Data Generation With Semantic Context," Proceedings of the Aaai Conference on Artificial Intelligence, vol. 38, no. 14, pp. 15996-16004, 2024, doi: 10.1609/aaai.v38i14.29531.
[13] T. Javidi, K. Nakhaei, H. Nakhaei, and M. Gholamzadeh, "Presentation of an Artificial Intelligence-Based Auditing Services Model with an Emphasis on Customer Trust: Theme Analysis and Single-Layer Perceptron Neural Network," Business, Marketing, and Finance Open, pp. 1-13, 2025.
[14] A. El-Fargani and M. El-Hassan, "Data-Driven Financial Fraud Detection Using Hybrid Artificial and Quantum Intelligence," Journal of Open Innovation: Technology, Market, and Complexity, vol. 11, no. 4, p. 100581, 2025.
[15] M. Bakhshi, "The scope of artificial intelligence in marketing: How AI impacts industries from financial services to retailing," Journal of Marketing Management, vol. 65, pp. 151-168, 2025.
[16] R. Sowmya, "Ai-Driven Consumer Behavior Analysis for Optimizing Marketing and Financial Strategies With Machine Learning," Journal of Information Systems Engineering & Management, vol. 10, no. 30s, pp. 357-363, 2025, doi: 10.52783/jisem.v10i30s.4842.
[17] A. Heydari Daloui, M. Vahdati, H. Mohebbi, and N. Bagherpour, "Predictability of the Tehran Stock Exchange total index using a hybrid machine learning approach: Analysis of market efficiency and the importance of influential variables," (in Persian), Financial Research, vol. 28, no. 2, pp. 464-493, 2026, doi: 10.22059/frj.2025.394747.1007738.
Downloads
Publication Timeline
- Submitted
- Revised
- Accepted
Issue
Section
License
Copyright (c) 2025 Seyedeh Soudabeh Jafari (Author); Mirfeiz Fallah Shams; Hamidreza Kordlouie, Fereydoun Rahnamae Roodposhti (Author)

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.