Credit Card Default Prediction: An Empirical Analysis on Predictive Performance Using Statistical and Machine Learning Methods

This article compares the predictive capabilities of six models, namely, linear discriminant analysis (LDA), logistic regression (LR), support vector machine (SVM), XGBoost, random forest (RF), and deep neural network (DNN), to predict the default behavior of credit card holders in Taiwan using data from the UCI machine learning database. The Python programming language was used for data analysis. Statistical methods were compared with machine learning algorithms using the confusion matrix measured in metric terms of prediction accuracy, sensitivity, specificity, precision, G-mean, F1 score, ROC, and AUC. The dataset contained 30,000 credit card users’ information, with 6636 default observations and 23,364 nondefault cases. The study results found that modern machine learning methods outperformed traditional statistical methods in terms of predictive performance measured by the F1 score, G-mean, and AUC. Traditional methods like logistic regression were marginally better than linear discriminant analysis and support vector machines in terms of the predictive performance measured by the area under the receiver operating characteristic curve. In the modern machine learning methods, deep neural network was better in the predictive performance metrics when compared with XGBoost and random forest methods.