Research On Credit Risk Assessment Of Listed Companies Based On Improved Rough Set And Support Vector Machines

Credit has a very important position and function in the modern market economic system,that It is the soul of the market economy and Its credit risk is one of the main risks of the enterprise.Listed company is as a representative of the enterprise,that its credit risk problem is of great practical significance in securities market regulation,investor benefit protection,and credit institution risk control.Moreover,as a kind of social supervision means to prevent corporate credit risk,credit risk assessment is the inevitable outcome of the development of market economy to a certain stage,through a series of theoretical research and practice of developed countries has also proved its importance and the promotion effect of the market economy.This paper is to study the problem of the credit risk assessment of listed companies,including the following aspects research content :First of all,this article from the credit risk of research background,research significance,Research Status at Home and Abroad its research of the main content and method of the four aspects of the elaboration.Combined with the domestic actual situation,and domestic and international credit risk assessment of the status quo of analysis and summary,clarified the selection listed companies as a research object and the cause of its credit risk produced.Also pointed out that China's current credit risk assessment of the existing problems.Secondly,based on the theory of improved Rough set(Rough set)attribute reduction and Stepwise regression on the factor selection theory,the paper the study of the construction of credit risk assessment index system of listed companies.This paper mainly uses the Rough set attribute reduction theory to The evaluation index set is reduced.This paper mainly uses the Rough set attribute reduction theory to The evaluation index set is reduced.In the process of the attribute reduction,when appear the index of same importance and that can not being reduced,use the stepwiseregression theory further reduced to construct a fair,objective,Scientific the evaluation index system.And then based on Financial Data of Listed Companies,use the AHP to determine the weight of indicators,express the financial indicators of listed companies of explaination the ability.Finally,based on the theory of density clustering with support vector machine,this paper makes a further study on the evaluation of credit risk of listed companies.It mainly use by the combining the principle of density clustering and the principle of support vector machine construct The credit risk assessment model of listed companies.Combined with the financial data of listed companies to conduct empirical research,get a predictive validity better evaluation model.