基于Fisher判别的小型工业企业债信评级模型及实证

›› 2018, Vol. 30 ›› Issue (3): 15-28.

基于Fisher判别的小型工业企业债信评级模型及实证

潘明道, 周颖, 迟国泰, 孟斌

大连理工大学管理与经济学部, 大连 116024

收稿日期:2016-06-17 出版日期:2018-03-28 发布日期:2018-03-26
通讯作者: 周颖(通讯作者),大连理工大学管理与经济学部副教授,硕士生导师,博士
作者简介:潘明道,大连理工大学管理与经济学部,博士;迟国泰,大连理工大学管理与经济学部教授,博士生导师,博士;孟斌,大连理工大学管理与经济学部,博士
基金资助:
国家社科基金项目（16BTJ017）；辽宁经济社会发展重点课题（2015lslktzdian-05）；教育部科学技术研究项目（2011-10）；中国银监会银行业信息科技风险管理项目（2012-4-005）；中国邮政储蓄银行总行小额贷款信用风险评价与贷款定价资助项目（2009-07）；大连银行小企业信用风险评级系统与贷款定价项目（2012-01）。

Small Industrial Enterprises' Credit Rating Model and Empirical Analysis Based on Fisher Discriminant

Pan Mingdao, Zhou Ying, Chi Guotai, Meng Bin

Faculty of Management and Economics, Dalian University of Technology, Dalian 116024

Received:2016-06-17 Online:2018-03-28 Published:2018-03-26

摘要/Abstract

摘要：

本文以中国某商业银行1814个小型工业企业贷款客户数据为样本，根据指标对违约状态鉴别精度的影响程度进行第一次筛选，保证遴选出的指标对违约状态鉴别能力都有显著影响；根据准则内相关分析进行第二次筛选，避免遴选出的指标反映信息重复，构建了一套能显著区分小型工业企业违约状态的评级体系。本文的创新与特色：一是根据有、无特定指标两种状态下、Fisher判别对违约状态鉴别精度的提高或降低，反映特定指标对违约状态的影响程度，剔除对违约状态的判别精度没有影响或有降低影响的指标，保留可以显著提高违约状态判别精度的指标，完善了现有研究遴选指标的标准与违约状态无关的不足。二是在相关系数大于0.7的两个指标中，根据对Fisher判别精度影响程度越大、这个指标区分违约状态能力越强的思路剔除对Fisher判别精度影响程度较小的指标，避免了现有研究在剔除冗余指标时、对违约状态影响大的指标可能被误删的不足。

关键词: 工业企业, 小型企业, 债信评级, Fisher判别

Abstract:

According to the influence degree of indexes on the identification precision of default state, this study conducts the first screening to ensure that the selected indexes have a significant effect to the identification ability of default state. According to correlation analysis within guidelines, this study conducts the second screening to keep the selected indexes from reflecting the same information. This study builds a set of rating system which can significantly distinguish the default state of small industrial enterprises. This paper makes an innovative and characteristic attempt in two aspects. Firstly, according to the condition either with or without specific indicators, effects of using Fisher discriminant on identification accuracy of default state, it reflects the influence of specific indicators on the default state. Secondly, with regards to the two indicators with correlation coefficient above 0.7, this paper, based on the thought that the greater impact on the Fisher discriminant accuracy, the stronger ability of the index to distinguish default state, accurately removes the indexes that have smaller influence on Fisher discriminant.

Key words: industrial corporate, small corporate, credit rating, fisher discriminant

潘明道, 周颖, 迟国泰, 孟斌. 基于Fisher判别的小型工业企业债信评级模型及实证[J]. , 2018, 30(3): 15-28.

Pan Mingdao, Zhou Ying, Chi Guotai, Meng Bin. Small Industrial Enterprises' Credit Rating Model and Empirical Analysis Based on Fisher Discriminant[J]. , 2018, 30(3): 15-28.

参考文献

[1] Stefan H. Credit Rating Dynamics and Competition[J]. Journal of Banking & Finance, 2014,49:100-112
[2] Michael B., Waseem L. Split Ratings and Differences in Corporate Credit Rating Policy Metween Moody's and Standard & Poor's[J]. Financial Review, 2014,49(4):713-734
[3] Silvia A., Sebastiano M. The Financing of Innovative SMEs:A Multicriteria Credit Rating Model[J]. European Journal of Operational Research, 2015,244(2):540-554
[4] 胡毅,王珏,杨晓光. 基于面板Logit模型的银行客户贷款违约风险预警研究[J]. 系统工程理论与实践, 2015,35(7):1752-1759
[5] 迟国泰,潘明道,程砚秋. 基于综合判别能力的农户小额贷款信用评价模型[J]. 管理评论, 2015,27(6):42-57
[6] 王颖,聂广礼,石勇. 基于信用评分模型的我国商业银行客户违约概率研究[J]. 管理评论, 2012,24(2):78-87
[7] Yu C., Xiao S. Y. A Sustainable Financing Credit Rating Model for China's Small and Medium Sized Enterprises[J]. Mathematical Problems in Engineering, 2014,10:80-85
[8] 杜永强,迟国泰. 基于指标甄别的绿色产业评价指标体系构建[J]. 科研管理, 2015,39(9):119-127
[9] 迟国泰,张亚京,石宝峰. 基于Probit回归的小企业债信评级模型及实证[J]. 管理科学学报, 2016,19(6):136-156
[10] Yu R.S., Stanley L. E. Fuzzy Numbers in the Credit Rating of Enterprise Financial Condition[J]. Review of Quantitative Finance and Accounting, 2001,17(4):351-360
[11] 胡海青,张琅,张道宏. 供应链金融视角下的中小企业信用风险评估模型——基于SVM与BP神经网络的比较研究[J]. 管理评论, 2012,24(11):70-80
[12] Meng B., Chi G. T. New Combined Weighting Model Based on Maximizing the Difference in Evaluation Results and its Application[J]. Mathematical Problems in Engineering, 2015:1-9
[13] Sumon K. B., Vinh D., Ali M. K. Implications of Bank Ownership for the Credit Channel of Monetary Policy Transmission:Evidence From India[J]. Journal of Banking & Finance, 2011,35(9):2418-2428
[14] Silvia A., Sebastiano M. The Financing of Innovative SMEs:A Multicriteria Credit Rating Model[J]. European Journal of Operational Research, 2015,244(2):540-554
[15] 张洪祥,毛志忠. 基于多维时间序列的灰色模糊信用评价研究[J]. 管理科学学报, 2011,14(1):28-37
[16] Chen S. S., Chen H. Y., Chang C. C., et al. How do Sovereign Credit Rating Changes Affect Private Investment?[J]. Journal of Banking & Finance, 2013,37(12):4820-4833
[17] Meng B., Chi G. T., Gong L. L. Credit Evaluation of Small Enterprises Based on Indicators' Diversification Maximized Combination Weighting Model[J]. Journal of Information & Computational Science, 2015,12(3):1267-1279
[18] 李天恩,何桢. 基于主成分修整和线性判别分析的重叠故障识别[J]. 系统工程学报, 2012,27(5):712-718
[19] 姚冰湜,马琳,王雪莉,等. 高管团队职能异质性对企业绩效的影响:CEO权力的调节作用[J]. 中国软科学, 2015,(2):117-126
[20] Kevin K. D., Alexei C. Sample Size Methods for Constructing Confidence Intervals for The Intra-Class Correlation Coefficient[J]. Computational Statistics & Data Analysis, 2015,85(85):67-83
[21] 迟国泰,石宝峰. 基于信用等级与违约损失率匹配的信用评级系统与方法(No. 201210201461.6)[P]. 中华人民共和国国家发明专利, 2015
[22] 工业和信息化部、国家统计局、发展改革委、财政部.中小企业划型标准规定[EB/OL]. 工业和信息化部网站, http://www.gov.cn/zwgk/2011-07/04/content_1898747.htm, 2011
[23] 大连银行. 大连银行小企业信贷系统[DB]. 大连银行, 2013
[24] 盛骤,谢式千,潘乘毅. 概率论与数理统计[M]. 北京:高等教育出版社, 2001
[25] 大连理工大学迟国泰教授科研创新团队. 大连银行小企业债信评级和信用评级系统研究[R]. 大连:大连理工大学(第1版), 2015