Credit Default Risk Measurement Of Listed Companies Based On Logit-Kmv Model

Since the reform and opening up,bond market,as an important direct financing channel of China's financial market,has developed rapidly,gradually forming multitier markets and diversified bond varieties.As the rigid payment model of China's bond market is broken,default events in China's bond market have been frequent since the first defaulted bond appeared in 2014.By the end of 2019,368 bonds had defaulted in China,with the default amount reaching 312.2 billion yuan.Defaults on credit bonds are concentrated in 2018 and 2019,with as many as 229 bonds defaulting between 2018 and 2019 alone.With the increasingly serious phenomenon of credit bond default,it is of great significance for investors and regulators to construct a reasonable and accurate risk measurement model and make reasonable prediction of credit bond default risk.Based on 2018 and 2019 bond defaults is most serious,this paper chose the credit bond material breach during these two years of listed companies as research samples,and select the reasonable control samples,build the KMV model,the LOGIT model and a LOGIT-KMV model comparative analysis,index selection contains a sample company and contrasting the market index,profitability index,capital structure,solvency index,operation index and growth index of six dimensions,by default the comparison between sample and reference sample build empirical prediction model,the results show that:(1)the default distance calculated based on KMV can accurately reflect the gap between the default samples and the control samples.The default distance of most listed companies included in the default samples is lower than that of the listed companies in the control samples.(2)the traditional LOGIT model is built based on the financial index,the default probability has good prediction effect,which sales net interest rate,asset-liability ratio,current ratio,inventory turnover,total asset turnover,main business revenue growth rate and growth rate of total assets of the seven financial means is a good index of risk identification,build the LOGIT model forecast in the sample company default risk,the accuracy reached 84.6%,the forecast effect is higher than the KMV model,the effect is better;(3)the default distance calculated by the KMV model was incorporated into the LOGIT model.For the improvement of the traditional LOGIT model,the LOGIT-KMV model was constructed,and both the model fitting effect and the default prediction effect were significantly higher than the traditional LOGIT model.