Font Size: a A A

Measurement On Default Risk Of Credit Bonds Based On KMV-Random Forest Model

Posted on:2020-06-18Degree:MasterType:Thesis
Country:ChinaCandidate:Y N ZhuFull Text:PDF
GTID:2439330572471593Subject:Applied statistics
Abstract/Summary:PDF Full Text Request
Since 40 years ago,China’s economic development has made remarkable achievements.As an indispensable part of the national economy,the financial market provides a channel for Chinese enterprises and investors to finance.As an important financing tool for enterprises,the research on China’s credit bonds has practical reference function for benign competition,market standardization development and rational investment of investors.The first part analyzes the phenomenon of default of credit bonds in China’s bond market,The domestic and foreign literatures on default risk of credit bonds are studied.It is found that the KMV model’s default distance and random forest model have strong practical value in quantifying the credit risk of China’s listed companies’bonds through the analysis of the development sequence in stages.Therefore,the two models are creatively combined.In the second part,data of 39 weak and 78 normal companies listed on the Shanghai and Shenzhen stock exchanges in the three years from 2015 to 2017 were selected to form a research sample,and data were standardized in different industries.Firstly,according to the KMV model,the default distance is calculated.It is found that there is a significant gap between the bad group and the control group,indicating that the KMV default distance can play a significant role in the classification of normal companies and bad companies.In the third part,the random forest model is used to filter the traditional financial indicators,then the random forest model evaluation index system is obtained.Observing the effect of the hybrid model,and comparing with the LOGIT model,it is proved that the hybrid model has a better effect in the quantification of default risk.In addition,it was observed by deleting the KMV default distance index.It shows that the combination of KMV model and random forest model has theoretical and practical significance.Finally,this paper puts forward corresponding suggestions and countermeasures for the avoidance of listed company bond default risk from the perspective of regulatory authorities,listed companies and market participants.
Keywords/Search Tags:Credit Bond Default Risk, KMV Model, Random Forest Model, Default Distance
PDF Full Text Request
Related items