Research On Early Warning Of Default Risk Of Enterprise Bond Based On Entropy Weight-Grey Relational Method

Under the background of the continuous expansion of bond issuance scale,the systematic risks accumulated in China's bond market are gradually exposed.The sustained growth of corporate bond issuance scale and the pressure brought by economic downturn have led to the fact that our government is no longer able to provide a bottom-up guarantee for large-scale bonds defaults,and instead a large number of corporate bonds have substantive defaults.From 2014 to 2018,there were 58 cases of corporate bond default in China,with the amount of default amounting to 49.422 billion yuan.Under the background of the normal development of bond default in China,it is of great significance for the long-term and stable development of China's bond market to deeply study the measurement and prediction of default risk of corporate bonds and the influencing factors of corporate bond default.Based on the theory of entropy,grey relational analysis and LOGIT regression analysis,this paper constructs an early warning model of corporate bond default risk.This paper focuses on the construction of financial indicators and bond information indicators system,the selection of research samples,the selection of indicators system,the construction of LOGIT regression model,and the prediction validity of quantitative model.Based on the sample of corporate bonds defaulted from 2014 to 2018,322 corporate bonds were selected as the research sample.This paper constructs an index system covering the financial information of the issuer and bond information,studies the influencing factors of corporate bond default risk by using the method of entropy weight and grey correlation analysis,builds a LOGIT model based on the data of one year before the default of bonds,and makes an analysis of the validity of the LOGIT model by testing samples.The empirical results show that: first,the grey correlation of entropy weight.The analysis can effectively calculate the correlation degree of the influencing factors and determine the influencing factors which are similar to the development trend of the explained variables.Secondly,the default risk measurement model of LOGIT bonds can calculate the default probability of bonds,and the accuracy of the model is 97.53%.This model can identify the default risk of corporate bonds.