THE APPLICATION OF SVM AND BNN IN CREDIT RISK ANALYSIS

摘要

In this paper we have a study of credit risk analysis from the credit-rating angle and attempt to extend previous research in two directions. Firstly, we apply a relatively new learning algorithm, support vector machines (SVM), to the credit-rating prediction problem and expect to improve prediction accuracy by adopting this new algorithm. Secondly, we apply the results from previous research on neural network model interpretation to the credit-rating problem, and try to provide some insights about the credit-rating process through neural network models. Based on these results, we conducted a market comparative analysis on the differences of determining factors in the United States and China markets.