شماره ركورد كنفرانس :
4857
عنوان مقاله :
Credit risk of bank loan case study in Iran
پديدآورندگان :
VahidiAsgari Amirsalar amirsalar.vahidi@aut.ac.ir Amirkabir University of Technology , Ebrahimi Seyede Niloofar niloofarebrahimi@aut.ac.ir Amirkabir University of Technology , Salavati Erfan e-mail: erfan.salavahi@aut.ac.ir Amirkabir University of Technology
كليدواژه :
Non , performing loan , Credit risk , Data mining.
عنوان كنفرانس :
پنجمين كنفرانس ملي مهندسي مالي و بيم سنجي
چكيده فارسي :
One of the most common indicators that used to identify credit risk is the ratio of non-performing loans (NPL). One major issue with granting loans is whether the borrowers could ful ll their obligation or not.The purpose of the present study is to identify the factors affecting the NPL of a bank for the period 2013-2017. We test machine learning methods against traditional methods on a collected dataset and the results proves that Random forest method works better among other techniques such as logistic regression, SVM, ANN and etc. And also decision tree is used as a rule extraction method.
