A Study on Forecasting the Default Risk of Bond Based on XGboost Algorithm and Over-Sampling Method

China’s bond market is an emerging market. The number of defaults has been increasing in recent years, but the data set severely imbalanced. Based on financial total 6731 corporate issuers which 50 had defaulted, this paper uses XGboost algorithm and Over-sampling method named SMOTE to predict default issuers. results show that advantages over traditional processing imbalanced data, one effective methods deal with samples. Then, way risk