An empirical comparison of conventional techniques, neural networks and the three stage hybrid Adaptive Neuro Fuzzy Inference System (ANFIS) model for credit scoring analysis: The case of Turkish credit card data

摘要: Abstract The number of Non-Performing Loans has increased in recent years, paralleling the current financial crisis, thus increasing importance credit scoring models. This study proposes a three stage hybrid Adaptive Neuro Fuzzy Inference System model, which is based on statistical techniques and Fuzzy. proposed model’s performance was compared with conventional commonly utilized models are tested using 10-fold cross-validation process card data an international bank operating Turkey. Results demonstrate that model consistently performs better than Linear Discriminant Analysis, Logistic Regression Artificial Neural Network (ANN) approaches, terms average correct classification rate estimated misclassification cost. As ANN, learning ability; unlike does not stay black box. In interpretation independent variables may provide valuable information for bankers consumers, especially explanation why applications rejected.