Agent-oriented modeling for credit risk analysis

Subject Areas : Journal of Investment Knowledge

Homa Azizi ¹ , Mohammadali Rastgav ²

1 - M.A Student in Financial Engineering of Economic University
2 - Associate Professor of Economic University

Received: 2015-02-14 Accepted : 2015-05-04 Published : 2015-05-04

Keywords: credit risk, Probability of Default, Agent-based modeling, Learning algorithm,

Abstract :

The credit crisis in recent years has increased the focus on bank credit risk. This paper uses an agent based model (ABM) to investigate the impact of bankers’ credit decision actions on bank credit losses that are induced by lending to corporate clients. In this model, we assume one bank  give credit to corporate clients and divide corporate in two sectores: small and medium corporates and large corporates. The results show that credit decision actions have substantial effects on bank credit losses, thus implying that regulators should consider organizational factors as a complement to bank assets when assigning capital requirements to banks. The study also aims to point to a new area of application of ABMs for both researchers and practitioners.