Prediction of Financial Distress in Companies Listed on the Tehran Stock Exchange Using Support Vector Machines (SVM) Optimized by Metaheuristic Algorithms

Sabzevari, Hossein; Rostami, Kiana

doi:10.25843/JIFM.2025.8563.21599

Prediction of Financial Distress in Companies Listed on the Tehran Stock Exchange Using Support Vector Machines (SVM) Optimized by Metaheuristic Algorithms

Document Type : Original Article

Authors

Hossein Sabzevari

Kiana Rostami

M.A. in Financial Management, University of Isfahan, Isfahan, Iran

10.25843/JIFM.2025.8563.21599

Abstract

Financial distress is considered one of the most critical stages preceding corporate bankruptcy, and its timely identification can prevent substantial losses for investors, creditors, and other stakeholders. In recent years, machine learning techniques have been widely employed in financial distress prediction due to their strong capability to extract complex patterns from financial data. In this study, the Support Vector Machine (SVM) is utilized as one of the most effective classification algorithms. Given the sensitivity of SVM performance to its tuning parameters, metaheuristic algorithms are employed to optimize the model parameters. The primary objective of this research is to develop a hybrid model based on SVM and metaheuristic optimization algorithms to improve the accuracy of financial distress prediction for companies listed on the Tehran Stock Exchange (TSE). Previous studies have demonstrated that the application of metaheuristic algorithms, such as the Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Ant Colony Optimization (ACO), can significantly enhance the performance of SVM models and improve prediction accuracy.

Keywords

Financial Distress Prediction

Support Vector Machine (SVM)

Metaheuristic Algorithms

Particle Swarm Optimization

Tehran Stock Exchange