Predicting the Effects of the Financial Crisis, Public Sector Accounting, Corporate Governance Mechanism and Audit Quality on the Financial Performance of Companies Listed on the Tehran Stock Exchange: The Performance of Neural Networks and Machine Learning
Year: 2024
Type: article
Abstract: This research examines the impact of financial crisis, International Public Sector Accounting Standards (IPSAS), corporate governance mechanisms (CG) and audit quality (AQ) on the financial performance of listed banks in Tehran Stock Exchange.The statistical population includes 24 listed banks during two periods: 1390-1397 (pre-crisis) and 1398-1402 (post-crisis).The dependent variables are return on assets (ROA), return on equity (ROE) and net interest margin (NIM).The research methodology is based on a combined machine learning approach using Fuzzy Neural Network (FNN) and Convolutional Neural Network (CANN).The FNN model is used to model uncertainties, while CANN aims to improve the generalizability of the model.The Particle Swarm Optimization (PSO) algorithm is used for parameter optimization.Model validation is performed using a 5-fold cross-validation technique.The results indicate that the financial crisis had a significant impact on the factors affecting the financial performance of banks listed on the Tehran Stock Exchange.During the crisis period, the importance of IPSAS, AQ, and CG increased significantly, highlighting the need for financial transparency and enhanced supervision under crisis conditions.Variables such as board size, composition, CAP and FSIZE showed a stronger positive effect on financial performance, while the negative effect of LEV increased.The optimized ANN-GA model showed high predictive accuracy (95.3% for ROA, 94.1% for ROE and 94.8% for NIM).Sensitivity analysis showed that in the pre-crisis period, CG, FSIZE and IPSAS had the most significant influence, while in the crisis period, the model's sensitivity to LEV, BSIZE and AQ increased significantly.Cross-validation tests confirmed the stability and high predictive power of the model, with the Diebold-Mariano test showing the significant superiority of ANN-GA over linear and logistic regression models.These results highlight the high capability of the ANN-GA model in identifying and predicting changes in factors influencing bank financial performance before and after the financial crisis.By emphasizing the importance of flexibility in banks' managerial and financial strategies in changing economic conditions, the study highlights the need for special attention to CG, IPSAS and AQ to improve financial performance.In addition, this research demonstrates the effectiveness of a combined machine learning approach in analyzing and predicting bank financial performance as a basis for strategic decision making in the banking industry and the formulation of regulatory policies.
Author: Suhailah Ahmed Hussein
Institution: Urmia University
Language: English
Subfield: Strategy and Management
Domain: Social Sciences
Open Access status: bronze
OpenAlex ID: W4404143489