Proposing a Model for Feasibility Assessment of Utilizing Artificial Neural Networks Optimized by Metaheuristic Algorithms for Predicting Stock Prices of Companies during Crises in the Tehran Stock Exchange

Message:
Article Type:
Research/Original Article (دارای رتبه معتبر)
Abstract:
Currently, investing in the stock market constitutes a significant portion of the country's economy. Securities are considered a reliable tool for gaining the trust of investors and are associated with various levels of risk. This approach can gather small and dispersed investments that, on their own, cannot be effectively utilized and transform them into substantial financial resources for economic development. In stock markets, price fluctuations are highly sensitive, leading to regular analysis and monitoring of these changes. As a result, stock price prediction has gained significant importance for investors, enabling them to maximize their returns and assisting them in making informed investment decisions. In recent years, modern time-series prediction methods based on artificial intelligence and machine learning have advanced rapidly. Given the high value of this data for investment and stock price prediction, traditional data analysis methods face limitations in effectively learning from it. With technological advancements and the introduction of new techniques, such as neural networks and metaheuristic algorithms, the use of these methods for stock price prediction has seen significant growth. In this study, the capability of various models based on neural networks optimized by two algorithms, Harris Hawk Optimization (HHO) and Honey Badger Algorithm (HBA), in predicting the stock price trends of two companies, Iran Khodro and Isfahan Oil Refinery, in the Tehran Stock Exchange over the next 10 days, is examined. The results of these two algorithms are compared using metrics such as MSE, RMSE, MAE, RSE, and EVS. The findings of this research indicate that the HBA algorithm outperforms the HHO algorithm, with accuracies of 75% and 76% in predicting the stock prices of Iran Khodro and Isfahan Oil Refinery, respectively, compared to the HHO algorithm's accuracies of 73% and 67%.
Language:
Persian
Published:
Emergency Management, Volume:14 Issue: 1, 2025
Pages:
1 to 36
https://www.magiran.com/p2829233  
سامانه نویسندگان
از نویسنده(گان) این مقاله دعوت می‌کنیم در سایت ثبت‌نام کرده و این مقاله را به فهرست مقالات رزومه خود پیوست کنند. راهنما
مقالات دیگری از این نویسنده (گان)