基于KMV模型的上市公司信用风险度量应用研究
|
摘要
信用风险作为金融风险中最主要的风险之一,是我国金融市场面临的巨大挑战。尤其在我国加入WTO以后,我国信用风险度量管理的重要性与日俱增。伴随着经济全球一体化、金融市场的迅速发展、金融竞争日益激烈,目前我国采用的传统信用风险度量方法,已无法满足现在信用风险度量工作的需要。因而本文通过借鉴国外的先进度量模型,建立适用于我国信用风险度量模型。
本文首先介绍信用风险的定义及其特征,其次对国外信用风险度量方法分为传统方法和现代模型进行分类阐述。其中传统度量方法中介绍了专家法、信用评分法和神经网络法,分析这些方法存在的缺陷和不足;现代度量模型则主要介绍了在国际上已得到广泛应用研究的Credit Metrics模型、KMV模型、Credit Portfolio View模型和Credit Risk+模型,在对这些模型的理论方法、应用步骤、优缺点及适用性分析后,选择较适合我国实际情况的KMV模型进行研究。然后对其理论基础和方法进行具体描述和参数讨论,根据讨论结果对其违约点设置进行修改。由于受到特殊处理的上市公司基本没有长期负债,非ST股的上市公司的长期负债在资产负债表中占有比例一般较低,且长期负债的偿债压力远低于短期负债,再加上KMV模型一般设定度量信用风险的有效期限为1年,因而考虑以短期负债为影响因素对违约点进行修改。经过以ST股和蓝筹股构成的开发样本和次新股为检验样本的研究发现,违约点在短期负债的60%处KMV模型最为有效合理。最后本文总结了工作内容和本论文存在的不足,并提出我国使用KMV模型度量信用风险的进一步研究方向。
Credit risk as one of the most significant financial risks is a great challenge faced by the financial market of our country. Especially after China’s accession to WTO, measurement and management of China's credit risk becomes more and more obvious. With economic globalization, rapid development of financial markets as well as increasingly fierce financial competition, presently the traditional measurement of credit risk methods adopted by China has been unable to meet the current needs of the work of credit risk measurement. Therefore, this paper is to develop a credit risk measurement model applicable to our country by drawing on the advanced measurement models internationally.
This paper first introduced the definition and characteristics of credit risk, and then divided the foreign credit risk measurement methods into traditional methods and modern models to describe, in which the expert method, the credit scoring method and the neural network method were described in the traditional methods of measurement as well as shortcomings and deficiencies in these methods; in the modern models, it was mainly to introduce Credit Metrics model, KMV model, Credit Portfolio View model and the Credit Risk + model, all of which have been widely applied and studied internationally, and choose KMV model, more suitable for China's actual situation to research after analysis of the theoretical approach, application steps, advantages and disadvantages and applicability of these models. Then, its theoretical basis and methods were specifically described and discussed on parameters, and according to results of the discussion setting of its default point was modified. As the listed companies specially handled basically have no long-term liabilities, long-term liabilities of the non-ST listed company account for a low e proportion generally in the balance sheet, and the paying pressure of long-term liabilities is far lower than short-term liabilities, coupled by one year valid generally set for the credit risk measurement of KMV model, thus, the default point was modified by considering the short-term liabilities as affecting factors. After studies of the development sample constituted by blue-chip stocks and ST stocks and the test sample constituted by sub-new stocks, it can be found that KMV model is most effective and reasonable when the default point is at the 60% position of short-term liabilities. Finally, this paper summarized the work contents and shortcomings of this paper and proposed the direction for further research on China’s application of the KMV model for measuring credit risk.
本文首先介绍信用风险的定义及其特征,其次对国外信用风险度量方法分为传统方法和现代模型进行分类阐述。其中传统度量方法中介绍了专家法、信用评分法和神经网络法,分析这些方法存在的缺陷和不足;现代度量模型则主要介绍了在国际上已得到广泛应用研究的Credit Metrics模型、KMV模型、Credit Portfolio View模型和Credit Risk+模型,在对这些模型的理论方法、应用步骤、优缺点及适用性分析后,选择较适合我国实际情况的KMV模型进行研究。然后对其理论基础和方法进行具体描述和参数讨论,根据讨论结果对其违约点设置进行修改。由于受到特殊处理的上市公司基本没有长期负债,非ST股的上市公司的长期负债在资产负债表中占有比例一般较低,且长期负债的偿债压力远低于短期负债,再加上KMV模型一般设定度量信用风险的有效期限为1年,因而考虑以短期负债为影响因素对违约点进行修改。经过以ST股和蓝筹股构成的开发样本和次新股为检验样本的研究发现,违约点在短期负债的60%处KMV模型最为有效合理。最后本文总结了工作内容和本论文存在的不足,并提出我国使用KMV模型度量信用风险的进一步研究方向。
Credit risk as one of the most significant financial risks is a great challenge faced by the financial market of our country. Especially after China’s accession to WTO, measurement and management of China's credit risk becomes more and more obvious. With economic globalization, rapid development of financial markets as well as increasingly fierce financial competition, presently the traditional measurement of credit risk methods adopted by China has been unable to meet the current needs of the work of credit risk measurement. Therefore, this paper is to develop a credit risk measurement model applicable to our country by drawing on the advanced measurement models internationally.
This paper first introduced the definition and characteristics of credit risk, and then divided the foreign credit risk measurement methods into traditional methods and modern models to describe, in which the expert method, the credit scoring method and the neural network method were described in the traditional methods of measurement as well as shortcomings and deficiencies in these methods; in the modern models, it was mainly to introduce Credit Metrics model, KMV model, Credit Portfolio View model and the Credit Risk + model, all of which have been widely applied and studied internationally, and choose KMV model, more suitable for China's actual situation to research after analysis of the theoretical approach, application steps, advantages and disadvantages and applicability of these models. Then, its theoretical basis and methods were specifically described and discussed on parameters, and according to results of the discussion setting of its default point was modified. As the listed companies specially handled basically have no long-term liabilities, long-term liabilities of the non-ST listed company account for a low e proportion generally in the balance sheet, and the paying pressure of long-term liabilities is far lower than short-term liabilities, coupled by one year valid generally set for the credit risk measurement of KMV model, thus, the default point was modified by considering the short-term liabilities as affecting factors. After studies of the development sample constituted by blue-chip stocks and ST stocks and the test sample constituted by sub-new stocks, it can be found that KMV model is most effective and reasonable when the default point is at the 60% position of short-term liabilities. Finally, this paper summarized the work contents and shortcomings of this paper and proposed the direction for further research on China’s application of the KMV model for measuring credit risk.
引文
[1]薛峰.银行信用风险分析[M].中国经济出版社,1995:10-14.
[2]韩平,席酋民.经济转轨时期我国商业银行信贷风险的分析[J].当代经济科学[J]. 1999,(4):22-26
[3]刘振亚,姚文雄.现代商业银行的贷款管理[M].中国城市金融. 1997,(7):51-54.
[4]毛振华.完善信用评级制度,有效防范金融风险[J].金融研究,1998 ,(7):68-70.
[5] Basel Committee on Banking Supervision. The New Basel Capital Accord[EB/OL]. http://www.bis.org/,2003.
[6]徐一丁.现代商业银行信用风险管理[M].四川大学出版社,2001:20-56.
[7]中国银行业监督管理委员会.统计数据[DB/OL]. http://www.cbrc.gov.cn/, 2008.
[8]Altman. Financial Ratios,discriminant analysis and prediction of corporat bankruptcy[J]. Journal of finance,1968,(9):189-209.
[9] Altman E.L.. Financial Ratios:Discriminant Analysis and the Prediction of Corporate Bankruptcy[J]. Journal of Finance,1968,(23):189-209.
[10] Scott E. The Probability of Bankruptcy:A Comparison of Empirical Predictions and Theoretical Models[J]. Journal of Banking and Finance,1981,(9):317-344.
[11]Altman,Haldeman, Narayanan. ZETA Analysis:A New Model to Identify Bankruptcy Risk of Corporations[J]. Journal of Banking and Finace,1991,(10):210-221.
[12] Press S.J.,Wilson S..Choosing between Logistic Regression and Discriminant Analysis[J]. America Statistics Association,1978,(73):699-705.
[13] Martin D.. Early Warming of Bank Failure: A Logit Regression Approch[J]. Journal of Banking and Finance 1977,(1):249-276.
[14] West R.C.. A Factor-Analytic Approach to Bank Condition[J]. Journal of Banking and Finance,2000,(3):253-266.
[15] Odom,Sharda. A Neural Network for Bankruptcy Prediction: International Joint Conference on Neural Network[J]. New York:New York University Press,1990:163-168.
[16] Altman , Macro. Corporate distress diagnosis:comparisons using linear discriminant analysis and neural networks[J]. Journal of Banking and Finance,1994,505-529.
[17] Kerling M.,Podding K.. Neuronal Netzeinder Okonomie[J]. Munchen,1994,(9):124-136.
[18] Malhotra D.K.. Differentiating between Good Credits and Bad Credits Using Neuron-Fuzzy Systems[M]. Computing Artificial Intelligence and Information Technology,2002,(136):489-501.
[19] Coats P.,Pant L.. Recoganizing Financial Distress Patteins using a Neural Network Tool[M].Finacial Management,1993,365-376..
[20]黄芳芳.信用风险计量法简介[J].城市金融论坛,1999,20(3):9-10.
[21] Christopher Finger. Conditional approaches for Credit Metrics Portfolio Distribution’s Monitor[J]. Technical Document,1999,39(15):1433.
[22] Credit Suiss. Credit Risk+:A Credit Risk Management Framework[M]. Credit Suisse Financial Products,1997,1753.
[23] Black F.,Scholes M.. The pricing of options and corporate liabilities[J]. Journal of Political Economy,1973,(8):637-659.
[24] Robot C. Merton. On the Pricing of Corporate Debt: The Risk Structure of interest Rates[J]. Journal of Finance,1974,28:449-470.
[25] Altman E.L.,Saunders A.. Credit Risk Measurement-Developments over the Last 20 years[J]. Journal of Banking and Finance,1998(21),1721-1742.
[26] Anthony Saunders. Credit Risk Measurement:New Approaches to Value at Risk and Other Paradigms London[J].2001,9(4):89-93.
[27] Michel Crouhy,Dan Galai,Robert Mark. A Comparative Analysis of Current Credit Risk Model[J]. Journal of Banking and Finance.2000,(24):59-117.
[28]巴塞尔银行监管委员会.外部信用评级与内部信用评级体系[M].中国金融出版社,2005:112.
[29] John A. M.. A comment on market vs. Accounting-based mearsures of default risk[Z]. Moodys KMV Corporation,1993,(7):210-223..
[30] Jorge R. Sobehart,Sean Keenan,Benchmarking Quantitative Default Risk Models: A Validation Methodology[Z]. Moodys KMV Corporation,2000,(3):113-117.
[31] Matthew Kurbat,Irina Korablev. Methodology for Testing the Level of EDF Credit Measure[Z]. Moodys KMV Corporation,2002,(8):71-79.
[32] Peter Crodie , Jedff Bohn. Modeling Default Risk[Z]. Moodys KMV Corporation,2003,(10):224-230.
[33]王春峰,万海晖,张维.商业银行信用风险评估及其实证研究[J].管理科学学报,1998,(1):63-67.
[34]王春峰,万海晖,张维.基于神经网络技术的商业银行信用风险评估[J].系统工程理论与实践,1999,(9):23-25.
[35]陈静.上市公司财务恶化预测的实证分析[J].会计研究,1999,(4):31-38.
[36]邹新月,李汉通.运用典型多元判断分析法评估上市公司信用风险[J].统计与决策,2001,(7):21-22
[37]吴世农,卢贤义.我国上市公司财务困境的预测模型研究[J].经济研究,2001,(6):46-55.
[38]于立勇,詹捷辉.基于Logistic回归分析的违约概率预测研究[J].财经研究,2004,30(9):15-23.
[39]徐佳娜,西宝.基于AHP-ANN模型的商业银行信用风险评估[J].哈尔滨理工大学学报,2004,(3):94-98.
[40]王琼,陈金贤.信用风险定价方法与模型研究[J].现代财经,2002,(4):14-16.
[41]程鹏,吴冲锋.信用风险度量和管理方法研究[J].管理工程学报,2002,(1): 70-73.
[42]薛峰,关伟.上市公司信用风险度量的一种新方法[J].西北工业大学学报,2003,(9):38-42.
[43]杨星,张义强.中国上市公司信用风险管理实证研究[J].中国软科学,2004(1):43-47.
[44]易丹辉,吴建明.上市公司信用风险计量研究[J].统计与信息论坛,2004(6):8-11.
[45]郑茂.基于EDF模型的上市公司信用风险实证研究[J].管理工程学报,2005(3):151-153.
[46]韩立岩.基于模糊随机方法的公司违约风险预测研究[J].金融研究,2002(8):48-53.
[47]郑承利,韩立岩.基于模糊统计的公司违约风险预测[J].模糊系统与数学,2003(3):84-89.
[48]邓宇翔,鲁炜.中国上市公司的KMV模型违约点的推算及比较研究[D],2005.
[49]熊健夫,孟卫东.基于KMV模型的信用风险度量实证研究[D],2007.
[50]鲁炜,赵恒珩. KMV模型关系函数推测及其在中国股市的验证[J].运筹与管理,2003(6):43-48.
[51]张玲,杨贞柿. KMV模型在上市公司信用风险评价中的应用研究[J].系统工程,2004(11):84-89.
[52]董颖颖. KMV模型在我国证券市场的适用性分析及其改进[J].生产力研究,2004(8).
[53]施兵超,杨文泽.金融风险管理[M]. 2002:12.
[54]巴塞尔委员会.有效银行监管的核心原则[M]. 1997.
[55]赵晓菊.银行风险管理[M].上海财经大学出版社,1999:20.
[56]石晓军,陈殿左.信用治理——文化、流程与工具[M].机械工业出版社,2004:10.
[57]黄万才.商业银行业务经营风险管理研究[M].重庆出版处,2006,1-3.
[58] Coats P.,Fant L.. Recognizing financial distress patterns using a neural network toll[J]. Financial management,1993,142-155.
[59]张维,李玉霜.商业银行信用风险分析综述[J].管理科学学报1998,(3):20-26.
[60] Gordy M.B.. A Comparative Anatomy of Credit Model[J]. Jounral of Banking and Finance,2000(24):119-149.
[61] Crouhy M.,Mark R.. A Comparative Analysis of Current Credit Risk Models, www.defaultrisk.com.
[62]杨蕴石,徐枞巍.高级内部信用风险度量模型方法的比较[J].科技导报,2004(7):51-53.
[63]陶铄,杨晓光.商业银行内部信用评价的比较研究[J].管理评论,2002,(9):34-36.
[64]张凤龙.中国国有商业银行不良资产处置研究[M].吉林人民出版社,2006,7-10.
[65]李斌.商业银行信贷业务理论与实务[M].沈阳师范大学经济研究论丛,2006,1.
[66] Altman E.I.,Kishore V.,Varetto F.. Corporate distress diagnosis:comparisons using linear discriminant analysis and neural networks(the Italian Experience)[J]. Journal of Banking and Fiance,1994,505-529.
[67] Altman E.. Finance Ratios:Discriminant Analysis and Prediction ofCorporate Bankruptcy[J]. Journal of Finance,1968,(9):589-609.
[2]韩平,席酋民.经济转轨时期我国商业银行信贷风险的分析[J].当代经济科学[J]. 1999,(4):22-26
[3]刘振亚,姚文雄.现代商业银行的贷款管理[M].中国城市金融. 1997,(7):51-54.
[4]毛振华.完善信用评级制度,有效防范金融风险[J].金融研究,1998 ,(7):68-70.
[5] Basel Committee on Banking Supervision. The New Basel Capital Accord[EB/OL]. http://www.bis.org/,2003.
[6]徐一丁.现代商业银行信用风险管理[M].四川大学出版社,2001:20-56.
[7]中国银行业监督管理委员会.统计数据[DB/OL]. http://www.cbrc.gov.cn/, 2008.
[8]Altman. Financial Ratios,discriminant analysis and prediction of corporat bankruptcy[J]. Journal of finance,1968,(9):189-209.
[9] Altman E.L.. Financial Ratios:Discriminant Analysis and the Prediction of Corporate Bankruptcy[J]. Journal of Finance,1968,(23):189-209.
[10] Scott E. The Probability of Bankruptcy:A Comparison of Empirical Predictions and Theoretical Models[J]. Journal of Banking and Finance,1981,(9):317-344.
[11]Altman,Haldeman, Narayanan. ZETA Analysis:A New Model to Identify Bankruptcy Risk of Corporations[J]. Journal of Banking and Finace,1991,(10):210-221.
[12] Press S.J.,Wilson S..Choosing between Logistic Regression and Discriminant Analysis[J]. America Statistics Association,1978,(73):699-705.
[13] Martin D.. Early Warming of Bank Failure: A Logit Regression Approch[J]. Journal of Banking and Finance 1977,(1):249-276.
[14] West R.C.. A Factor-Analytic Approach to Bank Condition[J]. Journal of Banking and Finance,2000,(3):253-266.
[15] Odom,Sharda. A Neural Network for Bankruptcy Prediction: International Joint Conference on Neural Network[J]. New York:New York University Press,1990:163-168.
[16] Altman , Macro. Corporate distress diagnosis:comparisons using linear discriminant analysis and neural networks[J]. Journal of Banking and Finance,1994,505-529.
[17] Kerling M.,Podding K.. Neuronal Netzeinder Okonomie[J]. Munchen,1994,(9):124-136.
[18] Malhotra D.K.. Differentiating between Good Credits and Bad Credits Using Neuron-Fuzzy Systems[M]. Computing Artificial Intelligence and Information Technology,2002,(136):489-501.
[19] Coats P.,Pant L.. Recoganizing Financial Distress Patteins using a Neural Network Tool[M].Finacial Management,1993,365-376..
[20]黄芳芳.信用风险计量法简介[J].城市金融论坛,1999,20(3):9-10.
[21] Christopher Finger. Conditional approaches for Credit Metrics Portfolio Distribution’s Monitor[J]. Technical Document,1999,39(15):1433.
[22] Credit Suiss. Credit Risk+:A Credit Risk Management Framework[M]. Credit Suisse Financial Products,1997,1753.
[23] Black F.,Scholes M.. The pricing of options and corporate liabilities[J]. Journal of Political Economy,1973,(8):637-659.
[24] Robot C. Merton. On the Pricing of Corporate Debt: The Risk Structure of interest Rates[J]. Journal of Finance,1974,28:449-470.
[25] Altman E.L.,Saunders A.. Credit Risk Measurement-Developments over the Last 20 years[J]. Journal of Banking and Finance,1998(21),1721-1742.
[26] Anthony Saunders. Credit Risk Measurement:New Approaches to Value at Risk and Other Paradigms London[J].2001,9(4):89-93.
[27] Michel Crouhy,Dan Galai,Robert Mark. A Comparative Analysis of Current Credit Risk Model[J]. Journal of Banking and Finance.2000,(24):59-117.
[28]巴塞尔银行监管委员会.外部信用评级与内部信用评级体系[M].中国金融出版社,2005:112.
[29] John A. M.. A comment on market vs. Accounting-based mearsures of default risk[Z]. Moodys KMV Corporation,1993,(7):210-223..
[30] Jorge R. Sobehart,Sean Keenan,Benchmarking Quantitative Default Risk Models: A Validation Methodology[Z]. Moodys KMV Corporation,2000,(3):113-117.
[31] Matthew Kurbat,Irina Korablev. Methodology for Testing the Level of EDF Credit Measure[Z]. Moodys KMV Corporation,2002,(8):71-79.
[32] Peter Crodie , Jedff Bohn. Modeling Default Risk[Z]. Moodys KMV Corporation,2003,(10):224-230.
[33]王春峰,万海晖,张维.商业银行信用风险评估及其实证研究[J].管理科学学报,1998,(1):63-67.
[34]王春峰,万海晖,张维.基于神经网络技术的商业银行信用风险评估[J].系统工程理论与实践,1999,(9):23-25.
[35]陈静.上市公司财务恶化预测的实证分析[J].会计研究,1999,(4):31-38.
[36]邹新月,李汉通.运用典型多元判断分析法评估上市公司信用风险[J].统计与决策,2001,(7):21-22
[37]吴世农,卢贤义.我国上市公司财务困境的预测模型研究[J].经济研究,2001,(6):46-55.
[38]于立勇,詹捷辉.基于Logistic回归分析的违约概率预测研究[J].财经研究,2004,30(9):15-23.
[39]徐佳娜,西宝.基于AHP-ANN模型的商业银行信用风险评估[J].哈尔滨理工大学学报,2004,(3):94-98.
[40]王琼,陈金贤.信用风险定价方法与模型研究[J].现代财经,2002,(4):14-16.
[41]程鹏,吴冲锋.信用风险度量和管理方法研究[J].管理工程学报,2002,(1): 70-73.
[42]薛峰,关伟.上市公司信用风险度量的一种新方法[J].西北工业大学学报,2003,(9):38-42.
[43]杨星,张义强.中国上市公司信用风险管理实证研究[J].中国软科学,2004(1):43-47.
[44]易丹辉,吴建明.上市公司信用风险计量研究[J].统计与信息论坛,2004(6):8-11.
[45]郑茂.基于EDF模型的上市公司信用风险实证研究[J].管理工程学报,2005(3):151-153.
[46]韩立岩.基于模糊随机方法的公司违约风险预测研究[J].金融研究,2002(8):48-53.
[47]郑承利,韩立岩.基于模糊统计的公司违约风险预测[J].模糊系统与数学,2003(3):84-89.
[48]邓宇翔,鲁炜.中国上市公司的KMV模型违约点的推算及比较研究[D],2005.
[49]熊健夫,孟卫东.基于KMV模型的信用风险度量实证研究[D],2007.
[50]鲁炜,赵恒珩. KMV模型关系函数推测及其在中国股市的验证[J].运筹与管理,2003(6):43-48.
[51]张玲,杨贞柿. KMV模型在上市公司信用风险评价中的应用研究[J].系统工程,2004(11):84-89.
[52]董颖颖. KMV模型在我国证券市场的适用性分析及其改进[J].生产力研究,2004(8).
[53]施兵超,杨文泽.金融风险管理[M]. 2002:12.
[54]巴塞尔委员会.有效银行监管的核心原则[M]. 1997.
[55]赵晓菊.银行风险管理[M].上海财经大学出版社,1999:20.
[56]石晓军,陈殿左.信用治理——文化、流程与工具[M].机械工业出版社,2004:10.
[57]黄万才.商业银行业务经营风险管理研究[M].重庆出版处,2006,1-3.
[58] Coats P.,Fant L.. Recognizing financial distress patterns using a neural network toll[J]. Financial management,1993,142-155.
[59]张维,李玉霜.商业银行信用风险分析综述[J].管理科学学报1998,(3):20-26.
[60] Gordy M.B.. A Comparative Anatomy of Credit Model[J]. Jounral of Banking and Finance,2000(24):119-149.
[61] Crouhy M.,Mark R.. A Comparative Analysis of Current Credit Risk Models, www.defaultrisk.com.
[62]杨蕴石,徐枞巍.高级内部信用风险度量模型方法的比较[J].科技导报,2004(7):51-53.
[63]陶铄,杨晓光.商业银行内部信用评价的比较研究[J].管理评论,2002,(9):34-36.
[64]张凤龙.中国国有商业银行不良资产处置研究[M].吉林人民出版社,2006,7-10.
[65]李斌.商业银行信贷业务理论与实务[M].沈阳师范大学经济研究论丛,2006,1.
[66] Altman E.I.,Kishore V.,Varetto F.. Corporate distress diagnosis:comparisons using linear discriminant analysis and neural networks(the Italian Experience)[J]. Journal of Banking and Fiance,1994,505-529.
[67] Altman E.. Finance Ratios:Discriminant Analysis and Prediction ofCorporate Bankruptcy[J]. Journal of Finance,1968,(9):589-609.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |