Personal Credit Evaluation Model Based On PROMETHEE-? Method And Its Application Research

One of the core contents of the New Basel Capital Accord is to establish a bank's internal risk rating system.The risks faced by banks include a variety of risks,of which credit risk is always the most concerned issue.Personal credit is the foundation of the social credit system.Active development of personal credit is conducive to improving the status quo of social credit.With the development of the economy and the improvement of national consumption levels,the demand for personal loans in China has continuously increased.Especially in recent years,Internet finance has been sought after by people,and various forms of personal credit business have emerged,which has further increased the scale of personal credit.The result is a high personal credit default rate and high financial risk,which jeopardizes the normal operation of financial institutions and poses a threat to the country's financial security.Therefore,only a comprehensive,effective and scientific assessment of credit risk and the ability of financial institutions to prevent and respond to risks can better enable capital to serve the society.This paper briefly analyzes the existing problems of personal credit evaluation methods,and introduces the PROMETHEE-II method to evaluate personal credit.The PROMETHEE-II method is a multi-attribute decision-making method.Compared with other personal credit evaluation methods,the PROMETHEE-II method introduces the advantage function to process the sample index information,which can reflect the advantages of different samples under a certain index.However,there are still some problems in the PROMETHEE-II method,which are mainly reflected in the following aspects:First,in the personal credit evaluation problem,the evaluation index needs to be determined.However,the PROMETHEE-II method does not determine the relevant criteria for the indicator.Then,the PROMETHEE-II method includes six dominant functions,five of which are linear or stepped and cannot cover all the cases in the actual problem.Finally,when using the PROMETHEE-II method for personal credit evaluation,the advantage function needs to be determined for the evaluation index.However,in the existing research,most of them are solved from the subjective point of view or by using expert review to solve the above problems,which leads to the instability of the evaluation results to some extent.Therefore,this paper has improved the PROMETHEE-II method for the above problems.Firstly,in view of the problem that there is no relevant standard for determining the index in the PROMETHEE-II method,this paper proposes an improved method for determining the Fisher score index,which gives different weights to the default and non-default samples of the F value in the original method.How to determine the evaluation indicators in the case of uneven sample.Then,for the PROMETHEE-II method,the dominant function can not cover some problems in the actual situation.This paper proposes two new dominant functions,which increases the adaptability of the dominant function in practical applications.Finally,aiming at the existing research or determining the dominant function according to the preferences of the decision makers or using the expert evaluation method to determine the dominant function,this paper calculates the evaluation accuracy of each set of dominant functions,and determines the overall accuracy of the results.The function solves the problem of determining the dominant function.The research in this paper consists of four parts.The first part is an introduction.First,the research background and significance are introduced.Then,the research status of the personal credit evaluation method and PROMETHEE-II method is reviewed.Finally,the main research contents and innovations of this paper are expounded.The second part is the construction of personal credit evaluation model.On the one hand,the steps of the traditional PROMETHEE-II method are elaborated in this paper.On the other hand,this article has improved the PROMETHEE-II method.The third part is an empirical study based on personal credit data.In this part,this paper uses Taiwan's credit card default data and Japanese personal credit approval data to conduct an empirical test on the improved PROMETHEE-II method.Moreover,the evaluation results of the empirical selection dominant function and other multi-attribute decision making methods(TOPSIS method,ELECTRE-III method)were compared.The fourth part is the conclusion.This part mainly summarizes the empirical research and further forecasts on this basis.This paper makes a comparative analysis of the empirical results and draws the following conclusions:First,it is effective to use the PROMETHEE-II method for personal credit evaluation.Secondly,compared with the subjective selection of the dominant function,the method of determining the dominant function based on the overall accuracy rate of the evaluation results is effective.Thirdly,compared with the original dominant function of PROMETHEE-? method,the exponential dominant function and multi-stage dominant function proposed in this paper can increase the applicability of the dominant function in the evaluation index.