基于计算实验金融的CCSME信用行为演化和传染效应研究

发布时间：2018-05-01 22:40

  本文选题：中小企业信用共同体 + 信用风险　； 参考：《电子科技大学》2015年博士论文

【摘要】：随着经济全球化的发展,中小企业已经成为全球经济中最为活跃的部分。中小企业不仅支持了经济社会的发展,而且对于推动科技创新也发挥着不可替代的作用。由于中小企业在金融资源获取方面没有与其贡献相匹配,因此,中小企业融资难问题成为世界各国积极探索和研究的课题。当前,我国正处于战略转型的关键历史时期,经济发展也进入到“新常态”,尽管政府和金融机构先后也出台了多项政策措施,力求提升中小企业的融资能力,但由于我国中小企业的自身经营管理和风险管控能力均较为薄弱,加之金融危机对于宏观经济环境的影响余波未平,中小企业的融资困境依然严峻。为了破解中小企业的融资瓶颈,“中小企业信用共同体”(Credit Community of Small and Medium-sized Enterprises,CCSME)融资模式应运而生。中小企业信用共同体(以下简称CCSME)通过横向监督、惩罚激励以及互相担保等机制,不仅有效降低了企业与银行之间的信息不对称和中小企业的融资成本,而且为银行提供了一些优质的贷款项目,实现了互利共赢的美好局面。在CCSME融资业务在全国范围内如火如荼地开展的同时,CCSME集体违约事件也频繁见诸于报端。CCSME缘何从防控风险的“保险锁”变成为信用风险传染的“导火索”?该问题成为学术界和业界关注的热点问题。迄今为止,国内外相关文献关于CCSME的探讨,大多限于农户或个体借贷者的联保贷款,对CCSME的研究也主要集中在可行性分析方面。由于CCSME整体的信用风险主要源于其内部成员企业的信用行为和成员企业之间的交互作用与影响,因此,CCSME成员企业的信用行为及其交互作用和CCSME内关联信用风险是影响CCSME整体信用风险的关键,然而,目前关于这方面的研究极为鲜见。基于此,本文针对上述问题展开深入研究,探索CCSME成员企业信用行为的演化规律和相互作用机理,并在此基础上探究CCSME内关联信用风险的传染效应以及CCSME的组织规模和关联结构对CCSME整体信用风险的影响机理。本文的研究不仅对CCSME成员企业的信用行为和CCSME内关联信用风险进行了刻画,而且对有效地降低CCSME的整体信用风险和完善CCSME融资模式提供重要的理论依据。本文的主要创新和特色在于紧密结合我国CCSME融资模式的实际情况,从行为科学和复杂系统视角出发,采用理论推导和模拟仿真相结合的方法,从微观和宏观两个维度对CCSME成员企业信用行为及其交互作用和CCSME内关联信用风险的传染效应展开研究,并为防控CCSME整体信用风险的爆发提供了相应的对策建议。本文的创新性工作体现在以下三个方面:第一,从行为科学视角出发,探究CCSME成员企业信用行为的演化机理。目前关于CCSME信用风险的研究,常常忽略了成员企业的信用行为所具有的有限理性特征,更没有反映成员企业之间的交互作用与影响。基于此,本文通过构建CCSME成员企业信用行为的演化博弈模型,对于成员企业之间信用行为的交互作用与影响进行了框架性描述与设定。进一步,分别基于前景理论(Prospect Theory,PT)和经验权重魅力值(Experience Weighted Atrraction,EWA)学习模型对于成员企业履约策略选择的偏好和信念进行刻画,讨论了在有限理性条件下,CCSME成员企业信用行为之间的交互作用对CCSME整体信用风险的影响机理。第二,从复杂系统视角出发,研究CCSME内关联信用风险的传染效应。鉴于CCSME内关联信用风险传染系统属于典型的复杂自适应系统,因此,本文借鉴复杂系统研究的相关理论和方法对其进行研究。利用元胞自动机(Celluar Automata,CA)模型对于CCSME内关联信用风险的传染机理和过程进行研究,并将CA模型嵌入适合度景观理论(Fitness Landscape Theory,FLT)对成员企业交互行为进行刻画,进而构建CANK模型探讨CCSME的规模和关联结构对CCSME整体信用风险的影响效应。最后,在针对CCSME典型模式所建立的复杂网络模型的基础上,引入极值动力学(Bak-Sneppen,BS)模型探讨了末位淘汰机制对于提高CCSME整体信用水平的作用。第三,将基于主体(Agent-based,AB)的建模思想和计算实验金融(Agent-based Computational Finance,ACF)的方法引入CCSME成员企业信用行为演化机理和关联信用风险传染效应的研究中,从CCSME成员企业的交互作用和影响出发,采用由下而上的建模方法,设计成员企业的信用行为,构建理论模型并进行模拟仿真,不仅有效地克服了传统方法的不足,而且与传统方法相互补充,提高了分析金融复杂系统的能力。另外,本文尝试将理论模型进行嵌套组合构建了复合模型。例如:第六章所构建的复合CA模型对于关联信用风险传染系统的刻画更加贴近现实;第七章所构建的CANK模型使两种模型的优势与特点均得到充分的发挥。本文将仿真模型进行有机组合后所构建的复合模型能够扬长避短,对于更好地刻画和模拟金融复杂系统具有一定的借鉴意义。
[Abstract]:With the development of economic globalization, small and medium enterprises have become the most active part of the global economy. Small and medium-sized enterprises not only support the development of economic society, but also play an irreplaceable role in promoting scientific and technological innovation. The problem of financing difficulty has become the subject of active exploration and research of all countries in the world. At present, China is in the key historical period of strategic transformation, and the economic development has also entered a "new normal". Although the government and financial institutions have also introduced a number of policies and measures to improve the financing capacity of small and medium-sized enterprises, but because of the small and medium-sized enterprises of our country The ability of personal management and risk management is weak, and the impact of the financial crisis on the macroeconomic environment remains uncommon, and the financing dilemma of small and medium-sized enterprises is still severe. In order to solve the financing bottleneck of small and medium-sized enterprises, "Credit Community of Small and Medium-sized Enterprises, CCSME" financing The model came into being. The SME credit community (hereinafter referred to as CCSME) through the lateral supervision, punishment incentive and mutual guarantee mechanism, not only effectively reduce the information asymmetry between enterprises and banks and the financing costs of small and medium-sized enterprises, but also provide some high-quality loan projects for the banks to achieve a better win-win situation. Situation. While the CCSME financing business is in full swing throughout the country, the CCSME collective default events are also frequently seen in the newspaper.CCSME why the "insurance lock" from the prevention and control risk is turned into a "fuse" for the contagion of credit risk? This problem has become a hot issue in the academic and industry. So far, domestic and foreign related The literature about CCSME is mostly limited to the loan of the farmers or individual borrowers, and the research on the CCSME is mainly concentrated on the feasibility analysis. Because the credit risk of the whole CCSME is mainly derived from the credit behavior of the internal members and the interaction and influence between the member enterprises, the credit line of the CCSME member enterprises It is the key to influence the overall credit risk of CCSME and its interaction and CCSME. However, the current research on this aspect is very rare. Based on this, this paper makes an in-depth study on the above problems to explore the evolution and interaction mechanism of the credit behavior of the CCSME member enterprises, and to explore the inside of CCSME on this basis. The effect of related credit risk and the influence mechanism of CCSME's organization scale and related structure on the overall credit risk of CCSME. This paper not only portrays the credit behavior of the CCSME member enterprises and the associated credit risk within the CCSME, but also provides the effective reduction of the whole credit risk of CCSME and the perfect CCSME financing mode. The main innovation and characteristic of this paper is to combine the actual situation of CCSME financing model in our country closely. From the perspective of behavioral science and complex system, the combination of theoretical deduction and Simulation simulation is adopted in the two dimensions of CCSME members' credit behavior and its interaction and CCSME inner customs from the micro and macro dimensions. The contagious effect of joint credit risk is studied and the corresponding countermeasures and suggestions are provided for the prevention and control of the CCSME overall credit risk. The innovative work of this paper is embodied in the following three aspects: first, from the perspective of behavioral science, this paper explores the mechanism of the credit behavior of the CCSME member enterprises. At present, the research on the credit risk of CCSME is often studied. It often ignores the limited rational characteristics of the credit behavior of the member enterprises, and does not reflect the interaction and influence of the member enterprises. Based on this, this paper constructs a framework of the evolution game model of the credit behavior of the CCSME member enterprises, and gives a framework description and establishment of the interaction and influence of the credit behavior among the member enterprises. Further, based on the Prospect Theory (PT) and the Experience Weighted Atrraction (EWA) learning model, the preference and belief of the performance strategy selection of the member enterprises are depicted, and the interaction of the credit behavior between the CCSME members is discussed under the limited rationality. Second, from the perspective of complex system, we study the contagious effect of associated credit risk within CCSME. In view of the CCSME related credit risk contagion system is a typical complex adaptive system, this paper uses the related theory and method of complex system research to study it by using the cellular automata (Celluar). Automata, CA) model studies the transmission mechanism and process of associated credit risk in CCSME, and embeds CA model into fitness landscape theory (Fitness Landscape Theory, FLT) to describe the interaction behavior of member enterprises, and then constructs CANK model to explore the effect of the scale of CCSME and the influence of association structure on the overall credit risk of CCSME. Finally, on the basis of the complex network model established in the typical CCSME model, the Bak-Sneppen (BS) model is introduced to explore the effect of the final elimination mechanism on improving the overall credit level of CCSME. Third, the modeling idea based on the Agent-based (AB) and the computational experimental finance (Agent-based Computational Finance) are used. In the study of the evolution mechanism of CCSME members' credit behavior and the contagious effect of associated credit risk, the method of ACF), starting from the interaction and influence of the CCSME member enterprises, uses a bottom-up modeling method to design the credit behavior of the member enterprises, constructs the theoretical model and carries out simulation simulation, not only effectively overcome the traditional side. The deficiency of the method and complementary to the traditional methods improve the ability to analyze the complex financial system. In addition, this paper tries to construct a composite model with the nested combination of the theoretical models. For example, the complex CA model constructed in the sixth chapter is more close to the reality of the related credit risk infection system; the seventh chapter constructs the model of the model. The advantages and characteristics of the two models are fully exerted. The composite model constructed by the organic combination of the simulation model in this paper can develop the advantages and avoid weaknesses, and has certain reference significance for the better depiction and Simulation of the complex financial system.

【学位授予单位】：电子科技大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：F276.3;F275
，

本文编号：1831310