租赁和商务服务业小企业的信用评价研究
发布时间：2018-12-29 18:19
【摘要】：我国的租赁和商务服务业小企业发展快速且数量众多,租赁和商务服务业小企业贷款难的问题一直是困扰这类企业发展的难题。由于现有的信用评价体系无法反映租赁和商务服务业小企业的信用评价特征,甚至绝大多数银行都没有建立租赁和商务服务业小企业的信用评价体系,因此我国租赁和商务服务业小企业的信用评价问题急需解决。 本论文由五章组成。第一章是绪论。第二章是基于显著性检验的租赁和商务服务业小企业信用评价指标体系构建。第三章是基于投影寻踪判别的租赁和商务服务业小企业信用评价模型。第四章是基于Copula-等分法的租赁和商务服务业小企业信用等级划分。第五章是结论与展望。 本论文的主要工作有三： (1)构建了租赁和商务服务业小企业信用评价指标体系。通过秩和检验、方差检验以及秩相关分析的三次组合方法筛选出能显著区别违约状态的信用评价指标体系。 通过秩和检验保留显著区分违约状态的信用评价指标；通过方差检验保留显著区分违约状态的信用评价指标；通过秩相关分析进一步删除区分违约状态能力弱的评价指标,建立了能够显著区分违约状态的租赁和商务服务业小企业信用评价指标体系。 (2)测算了租赁和商务服务业小企业的信用得分。通过违约企业与不违约企业最大分离的投影寻踪判别模型测算评价指标的权重,通过评价指标与权重的加权线性组合建立信用得分测算模型。 通过构造违约类样本投影点逼近负理想值、不违约类样本投影点逼近正理想值的投影寻踪判别模型,反映了违约企业样本与不违约企业样本差距越大则评价指标越重要的权重测算思路,解决了租赁和商务服务业小企业信用评价指标的权重测算问题。 (3)合理划分了租赁和商务服务业小企业的信用等级。通过Copula方法产生评级大样本数据,通过等分-动态调整法建立租赁和商务服务业小企业的合理评级模型。 通过信用得分、应还本息以及未还本息这三个变量的Copula联合分布函数模拟得到评级大样本,改变了小违约样本不能分级的现状,解决了小违约样本下如何建立信用等级越高而对应等级违约损失率越低的合理评级体系问题。实践中租赁和商务服务业小企业共113个样本,其中违约样本26个,现有小违约样本由于数量不足无法验证信用等级越高而对应等级违约损失率越低的评级体系。
[Abstract]:The small enterprises in leasing and business service industry are developing rapidly and in large quantities in China. The problem of loan difficulty for small enterprises in leasing and business service industry has always been a difficult problem for the development of this kind of enterprises. Because the existing credit evaluation system can not reflect the credit evaluation characteristics of small enterprises in leasing and business service industries, even most banks have not established credit evaluation systems for small enterprises in leasing and business service industries. Therefore, the credit evaluation of small enterprises in leasing and business service industry needs to be solved urgently. This thesis consists of five chapters. The first chapter is the introduction. The second chapter is the construction of credit evaluation index system of small enterprises in leasing and business service industry based on significance test. Chapter three is the credit evaluation model of small enterprises in leasing and business service industry based on projection pursuit. The fourth chapter is the classification of credit grade of small enterprises in leasing and business service industry based on Copula- and other methods. The fifth chapter is the conclusion and prospect. The main work of this paper is as follows: (1) the credit evaluation index system of small enterprises in leasing and business service industry is constructed. The credit evaluation index system which can distinguish the default state is selected by the three combination methods of rank sum test variance test and rank correlation analysis. The credit evaluation index which distinguishes the default state by rank sum test, the credit evaluation index by variance test, and the credit evaluation index which distinguishes the default state by the variance test. Based on rank correlation analysis, the index system of credit evaluation for small enterprises in leasing and business service industries, which can distinguish the state of default between leasing and business service enterprises, is further deleted. (2) the credit score of small enterprises in leasing and business service industry is calculated. The weight of the evaluation index is calculated by the projection pursuit discriminant model of the maximum separation between the defaulting enterprise and the non-defaulting enterprise, and the credit score calculation model is established by the weighted linear combination of the evaluation index and the weight. In this paper, a projection pursuit discriminant model is constructed, in which the projection point approximates the negative ideal value and the non-default sample projection point approximates the positive ideal value. It reflects that the bigger the gap between the sample of defaulting enterprise and the non-defaulting enterprise, the more important the weight of evaluation index is, and the problem of weight calculation of credit evaluation index of small enterprise in leasing and business service industry is solved. (3) the credit grade of small enterprises in leasing and business service industry is divided reasonably. The Copula method is used to generate large sample data, and the reasonable rating model for small enterprises in leasing and business service industries is established by the method of equipartition and dynamic adjustment. Through the credit score, the Copula joint distribution function simulation of the three variables, should pay back principal and interest and not repay principal and interest, obtained the large sample rating, which changed the current situation that small default sample can not be classified. It solves the problem of how to establish a reasonable rating system with higher credit grade and lower default loss rate of corresponding grade under small default sample. In practice, there are 113 samples of small enterprises in leasing and business service industry, of which 26 are default samples. Due to the insufficient number of small default samples, the higher the credit rating is, the lower the default loss rate of the corresponding grade is.
【学位授予单位】：大连理工大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：F276.3;F832.4
本文编号：2395201
[Abstract]:The small enterprises in leasing and business service industry are developing rapidly and in large quantities in China. The problem of loan difficulty for small enterprises in leasing and business service industry has always been a difficult problem for the development of this kind of enterprises. Because the existing credit evaluation system can not reflect the credit evaluation characteristics of small enterprises in leasing and business service industries, even most banks have not established credit evaluation systems for small enterprises in leasing and business service industries. Therefore, the credit evaluation of small enterprises in leasing and business service industry needs to be solved urgently. This thesis consists of five chapters. The first chapter is the introduction. The second chapter is the construction of credit evaluation index system of small enterprises in leasing and business service industry based on significance test. Chapter three is the credit evaluation model of small enterprises in leasing and business service industry based on projection pursuit. The fourth chapter is the classification of credit grade of small enterprises in leasing and business service industry based on Copula- and other methods. The fifth chapter is the conclusion and prospect. The main work of this paper is as follows: (1) the credit evaluation index system of small enterprises in leasing and business service industry is constructed. The credit evaluation index system which can distinguish the default state is selected by the three combination methods of rank sum test variance test and rank correlation analysis. The credit evaluation index which distinguishes the default state by rank sum test, the credit evaluation index by variance test, and the credit evaluation index which distinguishes the default state by the variance test. Based on rank correlation analysis, the index system of credit evaluation for small enterprises in leasing and business service industries, which can distinguish the state of default between leasing and business service enterprises, is further deleted. (2) the credit score of small enterprises in leasing and business service industry is calculated. The weight of the evaluation index is calculated by the projection pursuit discriminant model of the maximum separation between the defaulting enterprise and the non-defaulting enterprise, and the credit score calculation model is established by the weighted linear combination of the evaluation index and the weight. In this paper, a projection pursuit discriminant model is constructed, in which the projection point approximates the negative ideal value and the non-default sample projection point approximates the positive ideal value. It reflects that the bigger the gap between the sample of defaulting enterprise and the non-defaulting enterprise, the more important the weight of evaluation index is, and the problem of weight calculation of credit evaluation index of small enterprise in leasing and business service industry is solved. (3) the credit grade of small enterprises in leasing and business service industry is divided reasonably. The Copula method is used to generate large sample data, and the reasonable rating model for small enterprises in leasing and business service industries is established by the method of equipartition and dynamic adjustment. Through the credit score, the Copula joint distribution function simulation of the three variables, should pay back principal and interest and not repay principal and interest, obtained the large sample rating, which changed the current situation that small default sample can not be classified. It solves the problem of how to establish a reasonable rating system with higher credit grade and lower default loss rate of corresponding grade under small default sample. In practice, there are 113 samples of small enterprises in leasing and business service industry, of which 26 are default samples. Due to the insufficient number of small default samples, the higher the credit rating is, the lower the default loss rate of the corresponding grade is.
【学位授予单位】：大连理工大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：F276.3;F832.4
【参考文献】
相关期刊论文 前10条
1 刘淑莲;王真;赵建卫;;基于因子分析的上市公司信用评级应用研究[J];财经问题研究;2008年07期
2 马晓青;刘莉亚;胡乃红;王静;;小企业信用评估的模型构建与实证分析[J];财经研究;2012年05期
3 刘广斌;郭富贵;;中小型科技企业信用评价指标的构建与筛选[J];工业技术经济;2009年08期
4 邓超;胡威;唐莹;;基于拒绝推论的小企业信用评分模型研究[J];国际金融研究;2011年04期
5 王春峰,李汶华;商业银行信用风险评估:投影寻踪判别分析模型[J];管理工程学报;2000年02期
6 黎月红;徐艺心;;我国小企业贷款应用信用评分技术的研究[J];区域金融研究;2010年12期
7 迟国泰;章穗;齐菲;;小企业贷款信用评价模型及实证研究——基于最优组合赋权视角[J];财经问题研究;2012年09期
8 杨胜刚;夏唯;张磊;;信用缺失环境下的农户信用评估指标体系构建研究[J];财经理论与实践;2012年06期
9 郭文伟;陈泽鹏;钟明;;基于MLP神经网络构建小企业信用风险预警模型[J];财会月刊;2013年06期
10 周国强;王雪青;刘锐;;一种基于改进云模型的信用评价方法[J];系统工程;2013年04期
相关博士学位论文 前1条
1 程砚秋;基于支持向量机的农户小额贷款决策评价研究[D];大连理工大学;2011年
，本文编号：2395201
本文链接：https://www.wllwen.com/jingjilunwen/guojijinrong/2395201.html
