Multi-Class Imbalanced Corporate Bond Default Risk Prediction Based on the OVO-SMOTE-Adaboost Ensemble Model

Corporate bond default risk prediction is important for regulators, issuers and investors in the bond market. We propose a new approach for multi-class imbalanced corporate bond risk prediction based on the OVO-SMOTE-Adaboost ensemble model, which integrates the one-versus one (OVO) decomposition method, the synthetic minority over-sampling technique (SMOTE) and the Adaboost ensemble method. We categorize corporate bond default risk into three classes: very low default risk, relatively low default risk and high default risk, which is more scientific than the traditional two-class bond default risk, and carry out empirical experiments by respectively using DT, SVM, Logit and MDA as basic classifiers. Empirical results show that the prediction performance of the OVO-SMOTE-Adaboost (DT) model is overall better than the other three ensemble models such as OVO-SMOTE-Adaboost (SVM), OVO-SMOTE-Adaboost (Logit) and OVO-SMOTE-Adaboost (MDA). In addition, the OVO-SMOTE-Adaboost (DT) model greatly outperforms the OVO-SMOTE (DT) model, which is a single classifier model based on OVO and SMOTE without Adaboost. Therefore, the OVO-SMOTE-Adaboost (DT) model has satisfying performance of multi-class imbalanced corporate bond default risk prediction and is of great practical significance.

Economic Fluctuation, Local Government Bond Risk and Risk-Taking of City Commercial Banks

Countercyclical fiscal regulation can mitigate economic risk, but this is bound to increase the scale of local government debt during an economic downturn, and then spread risk to the banking sector, forming potential financial instability factors. We extracted the three most important variables in this process: economic fluctuation, local debt risk and bank risk-taking to build an econometric model and found that: (1) both economic fluctuations and local government bond risks have a significant impact on bank risk-taking, which is negatively correlated with local economic growth, while the increase of local government bond risks tends to increase bank risk-taking in the long run; (2) the impact of local government debt risk significantly increases the loans of city commercial banks flowing into the construction industry. Therefore, the impact of local government bond risk on city commercial banks is concentrated in the impact on their construction loans. This study has an important reference value for timely and moderate countercyclical regulation, preventing local debt risk from spreading to banks, constructing a sustainable local government−bank ecology, and promoting sustainable economic development.

Speculation, Sentiment, and Interest Rates

We compare the implications of speculation versus hedging channels for bond markets in heterogeneous agents’ economies. Treasuries command a significant risk premium when optimistic agents speculate by leveraging their positions using bonds. Disagreement drives a wedge between marginal agent versus econometrician beliefs (sentiment). When speculative demands dominate, the interaction between belief heterogeneity and sentiment helps rationalize several puzzling characteristics of Treasury markets. Empirically, we test model predictions and find that larger disagreement (i) lowers the risk-free rate, (ii) raises the slope of the yield curve, and (iii) with positive sentiment increases bond risk premia and makes its dynamics countercyclical. This paper was accepted by Karl Diether, finance.