Research Article
Improvement of Traditional Bond Default Identification Model Based on ESG Score
@INPROCEEDINGS{10.4108/eai.6-1-2023.2330338, author={Zhiyuan Bai and Nanyang Huang}, title={Improvement of Traditional Bond Default Identification Model Based on ESG Score}, proceedings={Proceedings of the 2nd International Conference on Big Data Economy and Digital Management, BDEDM 2023, January 6-8, 2023, Changsha, China}, publisher={EAI}, proceedings_a={BDEDM}, year={2023}, month={6}, keywords={corporate bonds default rate esg evaluation multiple linear regression stepwise regression}, doi={10.4108/eai.6-1-2023.2330338} }- Zhiyuan Bai
Nanyang Huang
Year: 2023
Improvement of Traditional Bond Default Identification Model Based on ESG Score
BDEDM
EAI
DOI: 10.4108/eai.6-1-2023.2330338
Abstract
The most important problem in the bond default field is whether the traditional model meets theoretical expectations well. To accurately identify bond default risk, this paper establishes a theory which introduces ESG scores into the traditional model of bond default identification based on some economic theory and tests the feasibility of the theory through empirical study methods. Has there been an improvement after incorporating ESG scores? After proving that incorporating ESG can truly predict reality well and have consistent practice by using the logistic regression method, this paper uses the CatBoost method in machine learning to solve an accurate prediction model. In the end, it was found that there is a significant positive relationship between ESG and bond default risk. After analysis, this paper draws the following results. First, ESG can help us better identify bond default risk. Second, when the company judges that its operating status is not good, it will whitewash the company through earnings management and public opinion to improve its ESG score. Third, the company will also default on bonds due to improper earnings management. To sum up, the establishment of ESG theory provides investors with a more precise method to assess the risk of corporate bonds.