Nasrin Motedayen - Faculty of Economics, Islamic Azad University, Tehran Central Branch, Tehran, Iran
Rafik Nazarian - Faculty of Economics, Islamic Azad University, Tehran Central Branch, Tehran, Iran
Marjan Damankeshideh - Faculty of Economics, Islamic Azad University, Tehran Central Branch, Tehran, Iran
Roya Seifipour - Faculty of Economics, Islamic Azad University, Tehran Central Branch, Tehran, Iran

The aim of this study is to identify and evaluate the factors that influence credit risk employing a multinomial logistic regression approach. For this purpose, in the first phase, indicators that affect credit risk assessment of natural customers were identified using documentation and library method. Then, the final data on the indicators were collected, including ۷۳۳۰ files of natural customers of Mellat Bank, and multinomial logistic regression was employed in studying the indicators of credit risk assessment of the bank’s natural customers in the four classes of timely receipt, overdue, deferred, and non-performing loans. The results of the estimated model show that the indicators of gender, loan value, age, installment interval, previous loan, occupation, loan repayment term, number of installments, quantity of each installment, loan extension, type of collateral, average balance, facility interest rate, type of facility, and education level have a significant impact on the credit risk of real customers.

Natural Customers, credit risk, Credit scoring, Multinomial Logistic Regression