Model for Predicting Bank Loan Default using XGBoost

Abstract

ABSTRACT Loan default prediction is one of the most important and critical problem faced by many banks and other financial institutions as it has a huge effect on their survival and profit. Many traditional methods exist for mining information about a loan application and have been greatly studied and applied in the past. These methods seem to be underperforming as there have been reported increases in the amount of bad loans and defaulters among many financial institutions. In this paper, gradient boosting algorithm called XGBoost was used for loan default prediction. The prediction is based on a loan data from a leading bank taking into consideration data sets from both the loan application and the demographic of the applicant. Similarly, important evaluation metrics such as Accuracy, Recall, precision, F1-Score and ROC area of the analysis were used. The paper provides an effective basis for loan credit approval in order to identify risky customers from a large number of loan applications using predictive modeling. The full utilization of this model will assist financial institutions in knowing a risking customer that may default in loan payment before lending.