Presenting a Model for measuring Credit Risk of Bank customers using Data Mining Approach Case:Tehran Tejarat Bank Customers

Presenting a Model for measuring Credit Risk of Bank customers using Data Mining Approach Case:Tehran Tejarat Bank CustomersFor managing credit risk, commercial banks use various scoring methodologies to evaluate the financial performance of client firms. In this study a parametric model (Logistic regression) and a nonparametric model (Classification and regression tree) used in order to conduct credit scoring models. These models designed by using 13 financial ratios as independent variables and gathering data from 282 companies and then compared in term of accuracy rate. These models specified the important variables which affect on classification and credit risk. Observed results detected that accuracy rate of nonparametric methods are competitive with parametric methods.or managing credit risk, commercial banks use various scoring methodologies to evaluate the financial performance of client firms. In this study a parametric model (Logistic regression) and a nonparametric model (Classification and regression tree) used in order to conduct credit scoring models. These models designed by using 13 financial ratios as independent variables and gathering data from 282 companies and then compared in term of accuracy rate. These models specified the important variables which affect on classification and credit risk. Observed results detected that accuracy rate of nonparametric methods are competitive with parametric methods.JEL classification: G21, E51