Document Type : Original Article
Authors
1
Department of Financial Engineering, Deh.C, Islamic Azad University, Isfahan, Iran.
2
Department of Accounting, ShK.C, Islamic Azad University, Shahrekord, Iran
3
Department of Management, Deh.C, Islamic Azad University, Isfahan, Iran.
10.22034/jik.2025.78226.4669
Abstract
With the expansion of the industry's role and its economic significance, preventing financial crises and corporate distress has become a key topic in financial management. The aim of this research is to propose a novel model based on deep learning and the Grey Wolf Optimization (GWO) algorithm for predicting financial distress. Using data from 160 companies listed on the Tehran Stock Exchange from 2012 to 2022, relevant financial indicators were identified. Subsequently, the proposed model's performance was evaluated by applying deep learning techniques and optimization algorithms.
Findings showed that the deep learning model based on the Grey Wolf algorithm, with an accuracy of 87%, outperformed the traditional artificial neural network model, which had an accuracy of 62%. This demonstrates the capability of artificial intelligence methods and metaheuristic algorithms in identifying complex patterns in financial data.
This research, by integrating multi-source data, advanced algorithms, and localizing the model for Iran's economic conditions, takes a novel step in providing preventive solutions and financial risk management. Practical applications of this model include identifying at-risk companies, supporting investment decision-making, and developing macro-level supportive policies.
Keywords: Financial distress, Deep learning, Grey Wolf Optimization, Financial prediction, Artificial intelligence