Research On Credit Rating Method Based On ELECTRE ? And Application

People will not succeed without credit.The industry will not prosper without credit.The country will fall without credit.China's credit problem has become increasingly prominent and needs to be solved.Since 2011,the non-performing loan balance and non-performing loan ratio of banks have shown a "double rise" trend.Under the pressure of non-performing loans,the steady operation of commercial banks also faces risks.The problem of credit also threatens the financial security of our country.The CPC Central Committee and the State Council attach great importance to this,and promote the introduction of a series of special plans and policy documents in order to speed up the construction of the integrity system.Credit rating is an effective way to deal with credit problems.However,China's credit rating industry has not developed well.Rating agencies lack of perfect and effective management and supervision,so it is difficult for credit agencies to give full play to their role.Deepening the theory of credit rating and improving the credit rating system are of great benefit to the development of China's financial market.Credit ratings are essentially multi-criteria decision-making issues.The ELECTRE method has been widely used since its introduction.Most of the multi-criteria decision-making methods cannot solve the Cask Effect problem in the credit rating process because of the decision-compensation.The ELECTRE ?method overcomes this problem by introducing a 'veto threshold.However,the subjectivity of threshold selection results in unstable rating results.The determination of threshold combinations in the existing literature mostly relies on expert review.Therefore,based on the existing literature,this paper proposes to search for the optimal threshold combination with the best rating results as the standard.In addition,the ELECTRE ? method needs to weight the rating indicators when calculating the consistency index.Therefore,the objective entropy method is chosen to carry out the index empowerment work.The first part of this article is an introduction.First of all,it introduces the background and significance of the topic.At this stage,China's credit problem is more prominent.The research in this paper is not only conducive to perfecting the credit rating barrel theory,making up for the ELECTRE ? method threshold to determine the vacancy;it can also alleviate the bank's "double-lift" pressure and reduce personal default risk.Secondly,the existing literature of the method is summarized and the research status of the method is reported.It has been found that the threshold combination selection of the ELECTRE ? method is still a difficulty.Finally,the innovations of this study are illustrated based on existing literature.The second part of this paper is method research.First,a brief introduction to the ELECTRE ? method.Secondly,the objective entropy method is selected to carry out the index weighting work,which solves the subjective and random problem of the index empowerment of the ELECTRE ? method.Finally,the optimal threshold combination is searched based on the discriminant precision and the wrong cost,which solves the problem of threshold setting subjectivity in the ELECTRE ?method.In this paper,by determining the five statistic of average absolute dispersion,mean,median,mode,and standard deviation,and setting the scale factor range of the threshold according to the existing literature,a threshold alternative set is constructed.Then use the constructed objective function as the criterion for rating results.By continuously narrowing the range of values of the threshold scale factor,the step size is reduced,and the search is iterative until the optimal threshold combination based on the statistic is determined.The third part of this paper is an empirical study based on FLECTRE ?.According to the method proposed in this paper,two open data sets,credit and bankruptcy,are used for empirical analysis.And compared with ELECTRE ?,ELECTRE ? method and other threshold combination determination methods.The analysis results show that the ELECTRE ? method is effective in solving the short board problem in the credit rating process.The feasibility of the statistic search optimal optimal threshold combination method proposed in this paper is verified.The rating results within the optimal interval of the threshold combination are robust.The recognition accuracy of the evaluation results using the optimal threshold combination method is higher than other similar methods.The fourth part of this paper is the conclusion and outlook.This section is mainly based on empirical research.This paper further confirms the effectiveness of the ELECTRE ? method in solving the short-board problem in the credit rating process.It also confirms the feasibility of the statistic-based search optimal threshold combination method proposed in this paper.And look forward to the shortcomings of this article.