Design and develop a model of credit risk in the banking system (Multilevel Modelling)

The observed crises in the banking systems of the countries are mainly due to inefficiencies in credit risk management. Due to the large volume of bank facilities, the risk of repaying them is a big challenge for banks. Therefore, in this paper, we have designed and developed a credit risk model in the banking system of the country by multilevel models. As a result, we use a logistic regression method to analyze the credit risk of legal clients of the country's banking system in a multi-level manner. The total number of observations used in the estimation of this model includes 5925 records of legal persons who have received Persian, Entrepreneur, Sine and Diet facilities. Determination of variables in the estimation of this model has been selected based on the step-by-step method. The results show that changing business at the first level will not change the credit risk of legal customers. Also, given that the coefficient of the third level is much higher than the first and second levels, displacement and change in the business within the second level, compared to the first and third levels, will have a much higher effect on the credit risk of legal customers. Therefore, the banking system should be more sensitive to the activities of the second level.