Predictive Analytics of Rural Bank Quality Credit

Credit is the main business of rural banks. Credit distribution cannot be separated from the risk of default by the debtor which has an impact on reducing credit quality. Worsening credit quality has the potential to reduce bank income because the bank's main income comes from loan interest income. Apart from that, worsening credit quality also has an impact on increasing the burden of provisions for losses on productive assets. One effort that can be made to minimize credit risk is to predict credit quality so that you can identify early the potential for a decline in credit quality. This research aims to obtain significant features that influence credit quality at Rural Banks and to predict credit quality classification at rural banks. The method used in this research is Ordinal Logistic Regression which will then be evaluated using ROC (Receiver Operating Characteristic) and AUC (Area Under Curve). The research results show that the best model for predicting credit quality uses all X variables, both credit information and debtor information, with an AUC value of 0.90 and a prediction accuracy of 93.44%.