Application of Artificial Intelligence in Tehran Stock Exchange

Application of Artificial Intelligence in Tehran Stock Exchange

Mohammad Hossein Abdorrahimian1 Ali Shayan Asl2

1) Assistant Professor, Department of Finance and Accounting, Faculty of Humanities, Meybod University, Meybod, Iran
2) M.Sc. Student in Financial Engineering, University of Science and Arts, Yazd, Iran

IOI Code : AAHC-ABHGDC
Publication : National Conference on Management, Commercialization, Knowledge-Based Economy and Investment for Production(impeb.ir)
Abstract :
    In recent years, the expansion of Artificial Intelligence (AI) and Deep Learning (DL) methodologies has opened new horizons in the forecasting and analysis of financial time-series data. This development is of paramount importance in capital markets, particularly within the Tehran Stock Exchange (TSE). Due to its high sensitivity to macroeconomic variables, political news, exchange rate fluctuations, investor sentiment, price limits, and minimum transaction volumes, the TSE exhibits highly complex and non-linear patterns. Under such conditions, relying solely on traditional technical and fundamental analysis methods is insufficient for effective investment decision-making and stock price prediction. This paper aims to review and analyze the applications of artificial intelligence in the Tehran Stock Exchange and to examine the most prominent algorithms utilized in both domestic and international research. Accordingly, algorithms such as Artificial Neural Networks (ANN), Support Vector Machines (SVM), deep learning models—specifically Long Short-Term Memory (LSTM)—and hybrid approaches for predicting stock prices, market indices, and market trends are investigated. The findings from the literature review indicate that deep learning-based models and hybrid methods possess superior accuracy in identifying non-linear market relationships compared to classical models. However, their efficacy remains contingent upon data quality, the selection of input variables, and the availability of adequate trading infrastructures. Finally, the most significant opportunities and challenges of implementing AI in the Tehran Stock Exchange, including data limitations, infrastructural deficiencies, and regulatory ambiguities, are discussed.

Keywords : artificial intelligence tehran stock exchange stock price prediction machine learning deep learning algorithmic trading