Applicability Analysis Of Modified KMV Model In A-Share Credit Risk Assessment

In recent years,the total amount of China's bond market has continued to expand.By the end of 2018,the total balance of China's bond market was 86 trillion yuan,and bond financing has become an important financing channel for enterprises.However,since 2018,with the gradual tightening of financial supervision and government supervision,there have been certain difficulties in the operation and financing of enterprises.Under the pressure of various factors,the largest concentration of bond defaults since the rigid redemption has occurred.There are 43 newly added default entities,including many large and medium-sized private enterprise listed companies.Because listed companies are usually large in scale and have many stakeholders,the losses and chain reactions caused by default are huge.Consequently,preventing and controlling the scope and systemic risks that may arise from the default of large and medium-sized listed companies should be one of the important tasks in the current battle to prevent and resolve major risks.In order to effectively prevent and control the credit risk problem in the context of the default of bond defaults,it is necessary to establish a dynamic and effective risk monitoring system to improve the ability of rating agencies to identify default risks in advance.Among the existing credit risk measurement methods,the KMV model is unique in that it only needs to utilize the dynamic data of the capital market and does not directly depend on the credit history data.Therefore,studying the applicability of the KMV model in China's securities market is of great value to help establish a sound and perfect local rating system.This paper revolves around the credit status of 15 new A-share listed companies that were newly defaulted in 2018.At the same time,15 "ST" companies and 15 companies with the subject rating of "AAA" were randomly selected among the A-share listed companies.The KMV model is used to empirically study the credit risk of three groups of listed companies.Through the calculation,analysis and comparison of the expected default probability of the three groups of samples,the applicability of the KMV model to assess the credit risk of A-share listed companies is studied.In addition,this paper introduces the PSO-KMV model optimized by particle swarm optimization algorithm,and uses the PSO-KMV model to treat the three groups of listed companies the same as before.The default distance and expected default probability of the three groups of listed companies under the KMV model and the PSO-KMV model are compared horizontally.This paper also conducts a targeted credit risk assessment for highly leveraged real estate companies,additionally.This paper finds that the KMV model can identify the credit risk differences of the substantive default group,ST group and AAA group.The default distance of the three groups of listed companies increases,while the expected probability of default decreases.The PSO-KMV model optimized by particle swarm optimization algorithm is more sensitive to the default risk of listed companies than the KMV model,so it can significantly improve the accuracy of credit risk prediction,especially for the default of the substantive default group.Under the KMV model,five real estate companies with a AAA rating and a large scale have consistent perform.The expected default probability is close to 1,indicating that the high leverage of the real estate industry implies a high risk of default.