Research On Enterprise Credit Rating Based On Ordered Logistic-KMV Combination Model

Since the reform and opening up,China's economic development has been rapid,and the development of the financial market has continued to be in line with international standards.The differences between the domestic and foreign markets have also gradually become clear.Especially since 2014,China's economic development has entered a new normal.By April 2018,the nominal value of China's fixed income market had exceeded RMB 8 trillion.There have been several recent major bond defaults?such as Huaxin bonds?.Credit rating has become an important basis for controlling investment risks and determining the interest rate of bonds.It has important research value.Whether it is from the point of view of debt-issuing companies or from the perspective of investors,research on corporate credit is a hot topic.Domestic and foreign scholars have been actively exploring the methods of corporate credit assessment,and have continuously developed and innovated in the application of models and the selection of variables.So far has developed a lot of corporate credit assessment methods,such as discriminant analysis,survival analysis,Logistic regression,etc.,there are artificial neural network methods,support vector machines and other artificial intelligence models.These methods have better properties for the measurement of credit risk,but there are still many insurmountable problems.This paper tries to find a model of credit risk measurement of listed companies suitable for China's market from the perspective of China's financial market.This paper attempts to innovate in the selection of models and explanatory variables.In terms of interpreting variables: Based on the financial variables,this paper adds indicators such as the size of the company and the industry category.In addition,macroeconomic development indicators such as regional GDP and per capita consumption index of the regions to which the company belongs are included in the model.In the aspect of model selection,this paper adopts a combination model to model,that is,the KMV model widely used by scholars and the ordered multi-category Logistic regression model are appropriately combined,and the combination model can overcome the shortcomings of a single model and give full play to each model.Theadvantages.The specific ideas of this paper are as follows: The KMV model is used to calculate the default distance indicators for each company in each year;this index is clustered using the K-means clustering method,and the company is divided into five ordered grades as the company's credit rating index.Finally,based on this indicator and the explanatory variables mentioned above,four orderly multi-classified Logistic regression models were established: using the financial and macroeconomic indicators one year ahead,a model Log₁ was established for the company's credit rating for that year,using two years in advance.Financial and macroeconomic indicators establish a model for the company's credit rating for that year.At the same time,in order to explore the influencing factors of changes in the corporate credit rating,models Log₃ and Log₄ are also established.Log₃ reflects the influence of financial and macroeconomic indicators one year in advance on the credit rating changes of the company in the current year.The model Log₄ reflects the impact of two years' financial and macroeconomic indicators on the credit rating changes of the company in the current year.As a result,it has been found that the level of corporate credit rating is affected by factors such as the company itself and the external environment,covering all aspects such as corporate solvency,profitability,operational capacity,development capability,company size,macroeconomic indicators,and industry indicators;The changes in corporate credit ratings are mainly affected by the company's own solvency and development capabilities.According to the research conclusion,this article puts forward from the perspective of the enterprise: If the enterprise wants to achieve a higher credit rating,it needs to constantly improve and enhance the capabilities of all parties.In addition,it should also pay more attention to the enterprise's ability to repay debt and develop its ability.