基于Copula-MKMV模型的供应链企业信用组合优化

发布时间：2018-06-07 10:02

本文选题：信用风险 + KMV模型　；参考：《上海师范大学》2014年硕士论文

【摘要】：随着供应链的不断发展，供应链突破了传统企业的边界，但供应链企业仍是相对独立的利益实体。所以供应链的管理范围变得更加广阔，因此在其组织和运营的过程中，也伴随着越来越多的风险。如何将各自的信用状况联系在一起获得整个系统的风险状况是进一步的问题。目前供应链企业风险研究多是从定性的角度来对风险进行评估，而且对信用风险的定量研究不多。而金融机构向供应链企业提供融资时需要确定的指标和数据来判断企业的投资风险、确定资产组合的比例，以保证收益最大化的同时风险最小。所以从整个系统的角度研宄供应链企业的信用风险更加迫切并有很好的实际意义。本文构建了一个供应链企业组合信用风险度量体系，，为商业银行等金融机构对供应链企业融资提供定性的参考。在求解的过程中采用GARCH模型修正后的KMV模型对供应链企业的风险进行度量，结果发现KMV模型可以充分地反映企业的风险水平，并且与经济环境密切相关。本文采用五元正态Copula函数、五元t-Copula函数、Gumbel-Copula函数、Clayton-Copula函数和Frank-Copula函数模型来度量整个供应链的风险水平，即联合违约距离。然后求山它们与经验Copula函数之间的平方欧式距离作为优劣的指标，得到平方欧式距离最小的模型一一五元t-Copula模型来作为本文样本的最优模型。在研究供应链贷款组合优化的过程中进行了充分的探讨，最终选择将组合权重这一约束条件放在各家企业违约距离的求解过程中来进行约束，通过随机数模拟产牛.大量的贷款权重组合进行求解，最终得到最优的贷款组合权里。
[Abstract]:With the continuous development of supply chain, supply chain breaks through the boundary of traditional enterprise, but supply chain enterprise is still a relatively independent benefit entity. Therefore, the scope of supply chain management becomes broader, so in the process of organization and operation, there are more and more risks. How to get the risk of the whole system is a further problem. At present, supply chain enterprise risk research is mostly from the qualitative point of view to evaluate the risk, and the quantitative study of credit risk is rare. When providing financing to supply chain enterprises, financial institutions need to determine the index and data to judge the investment risk and determine the proportion of the portfolio, so as to ensure the maximum return at the same time the minimum risk. Therefore, it is more urgent and practical to study the credit risk of supply chain enterprises from the point of view of the whole system. In this paper, a supply chain portfolio credit risk measurement system is constructed to provide a qualitative reference for commercial banks and other financial institutions to supply chain enterprise financing. In the process of solving, the modified KMV model of GARCH model is used to measure the risk of supply chain enterprises. The results show that the KMV model can fully reflect the risk level of enterprises and is closely related to the economic environment. In this paper, quaternion normal Copula function, quaternion t-Copula function, Clayton-Copula function and Frank-Copula function model are used to measure the risk level of the whole supply chain, that is, the joint default distance. Then the square Euclidean distance between them and the empirical Copula function is obtained as the index of superiority and inferiority and the model with the least square Euclidean distance is obtained as the optimal model of the sample in this paper. In the course of studying the optimization of supply chain loan portfolio, this paper makes a full discussion, and finally chooses to put the constraint condition of portfolio weight in the process of solving the default distance of each enterprise, and then simulates cattle production by random number. A large number of loan weight combinations are solved, and finally the optimal loan portfolio weight is obtained.
【学位授予单位】：上海师范大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F274;F224

【参考文献】