Research On Credit Evaluation For Small Enterprises Of Leasing And Business Services

China’s small enterprises of leasing and business services are numerous and fast-developing, but lacking of loans has always been a difficult problem to these enterprises. Because existing credit evaluation system does not reflect the credit evaluation’s characteristics of these enterprises, even the vast majority of banks have not been established credit evaluation system of these enterprises, credit evaluation for China’s small enterprises of leasing and business services is of an urgent problem.This dissertation consists of five chapters. The first chapter includes main concept definitions, existing relevant research, main research contents and methods, and so on. The second chapter establishes credit evaluation index system for small enterprises of leasing and business services based on significant test. The third chapter establishes credit evaluation model for small enterprises of leasing and business services based on projection pursuit discrimination. The fourth chapter obtains credit rating division for small enterprises of leasing and business services based on Copula-equal partition method. The fifth chapter includes conclusions and future perspectives.The main work of this dissertation is as follows:(1) This dissertation constructs credit evaluation index system for small enterprises of leasing and business services. This dissertation uses the triple combination method including rank sum test, variance test and rank correlation analysis to filter the credit evaluation index system which can significantly distinguish default state.This dissertation constructs credit evaluation index system for small enterprises of leasing and business services, which can significantly distinguish default state. First, retain credit evaluation indexes that can significantly distinguish default state through rank sum test. Second, choose credit evaluation index that can significantly distinguish default state by variance test. Third, delete credit evaluation index that cannot significantly distinguish default state using rank correlation analysis.(2) This dissertation measures credit scores for small enterprises of leasing and business services. Firstly, this dissertation uses projection pursuit discrimination model which can largestly separate default business and non-default business to measure the weight of the evaluation indicator. Secondly, this dissertation uses the weighted linear combination between the evaluation indicator and its weight to establish measurement model of the credit scores for small enterprises of leasing and business services.This dissertation constructs projection pursuit discrimination model, using projection point of default class sample to approach negative ideal value, and projection point of non-default class sample to reach positive ideal value. The model reflects weight calculation ideas that the more default sample differs from non-default sample, the more important the evaluation index is, thus solves weight measurement problem of the credit evaluation index.(3) This dissertation reasonably divides credit rating for small enterprises of leasing and business services. Firstly, this dissertation uses Copula method to generate large sample data of rating model for small enterprises of leasing and business services. Secondly, this dissertation uses combination method of equal partition and dynamic adjustment to establish reasonable rating model for small enterprises of leasing and business services.This dissertation uses Copula joint distribution function to simulate large sample through three rating variables, based on113samples for small enterprises of leasing and business services including26default samples. It develops small default sample credit rating theory and constructs reasonable rating system that satisfies order relations of the higher the credit rating the lower LGD. It improves the existing small default sample research that cannot undertake credit rating with narrow data samples.