Document Type : Research Paper
Faculty of Management, University of Tehran, Tehran, Iran
Faculty of Management, University of Tehran, Tehran, Iran.
Objective: One of the most important activities of banks is funds allocation and lending to different economic sectors. According to monetary and fiscal policies that are set by governments and the central bank, Banks allocate these resources to profitable and appropriate sectors. Controlling and reducing credit risk has always been considered as one of the effective factors in improving the lending process and consequently in the bank’s performance. Risk minimization along with profit maximization is a goal that banks always seek to achieve. Insufficient attention to the issue of loan returns and risk has led to the concentration of loans in certain economic sectors, which in itself has caused major problems for banks. The purpose of this study is to provide a model for risk measurement and optimization of a bank’s credit portfolio.
Methods: In this study, first, the credit risk of the loan portfolio of bank is investigated using the actuarial approach, and the value at risk is calculated as a risk criterion, then using the Perceptron neural network and according to the bank's limitations in lending, the optimal combination of credit portfolio will be determined. The sample which is used includes bank loans to 280 of its major customers in 4 sectors: industrial, trade and service, agriculture, and construction in 2013.
Results: The results show that the risk-adjusted return of the optimal portfolio is higher than the return of the current portfolio of the bank. In the optimal portfolio, the largest share is related to the agricultural sector, and the trade and service, construction and, industrial sectors are in the next categories, respectively. While in the current portfolio of the bank, the largest share of Loans is allocated to the industrial sector, followed by services and trade, agriculture, and construction, respectively.
Conclusion: Based on the results and confirmation of the research hypotheses, it can be concluded that the use of an actuarial model to determine credit risk and then optimization by the artificial neural network, leads to improving the process of optimizing the credit portfolio of banks.