Presenting the early warning model of financial systemic risk in Iran's financial market using the LSTM model
نویسندگان
1 Assistant Professor, Department of Islamic Economics, Faculty of Economic and Administrative Sciences, University of Qom, Qom, Iran
2 Assistant Professor, Faculty of Accounting and Management, Roudhen Unit, Islamic Azad University, Roudhen, Tehran, Iran
3 PhD student, Faculty of Accounting and Management, Roudhen Unit, Islamic Azad University, Roudhen, Tehran, Iran
4 Master of finance analystic, California state university long beach
doi
10.30495/ijfma.2024.77586.2115چکیده
The purpose of this article is to provide an early warning model of financial systemic risk in the financial market of Iran using the LSTM model. In this study, a long short-term memory (LSTM) approach was used to predict financial risk and yield changes in the country's capital market in the period of 2011-2023. In order to model the financial risk, the profitability of the banking, insurance and leasing industry and the total capital market index along with the exchange rate, interest rate, inflation rate and production variables were used. The designed model showed that it had a high power in predicting the fluctuations and occurrence of risk in the country's financial markets. In addition, the results obtained from the evaluation of the model on the test data were used to measure the performance of the system in generalizing the network training to the test stage and the ability of the model in predicting the efficiency of the financial industry and also as a warning system.
