中国金融系统风险测量与预警系统构建

发布时间：2018-05-24 05:45

本文选题：金融系统风险 + 预警系统　；参考：《复旦大学》2013年硕士论文

【摘要】：金融世界变幻莫测——每一天都存在新的金融产品、新的金融条例法规以及新的金融风险不断涌入金融系统；而对于金融市场迅猛发展的新兴国家中国而言,这些尤其值得重视。宏观金融系统的安全运行关系到整个国家和社会的稳定,如何找到蕴藏在金融系统、宏观经济系统中的金融风险因子,以及如何根据这些金融风险因子有效捕捉宏观金融系统风险信息并构建行之有效的风险预警系统可以对宏观金融风险的发生防患于未然,而这也正是本论文的主旨所在。论文的中心思想是分别构建两个模型对宏观金融系统风险程度进行测量和预测——即预警系统的构建。在国内和国际学术界,曾有多位知名学者在该领域研究做出卓越贡献,他们提出并试验了诸多模型,这些都将会在论文伊始的文献综述中逐一、简要回顾。本论文创新采用了因子分析进行宏观金融系统风险测量,进而又采用BP神经网络分析进行宏观金融系统风险预测及预警系统构建,两个模型相辅相成、相得益彰更是本论文的亮点之一。两个模型中,因子分析相对传统,可帮助学者更准确挖掘宏观金融系统的风险因子,同时实现输入变量的降维；而BP神经网络则是较为新兴的统计研究工具,将其运用于金融系统风险的测量和预警系统的构建在学术界更是首屈一指,其在非线性函数拟合中的卓越表现使其成为金融数据分析工具的不二之选。本论文是传统模型和新兴模型的结合体,但无论传统还是新兴,模型产生的测量和预测结果与现实贴切、令人信服。构建模型的目的在于更明晰地理解现实,然而模型仅仅是手段,而非目的；因此,基于模型的输出结果,论文进一步阐述宏观金融体系和宏观金融子体系的风险因子的经济意义,其测量和预测结果的实际意义和指导性作用。论文的结尾部分阐述了论文中所用模型的不完善之处,提出政策建议以增强宏观金融系统风险预警系统的有效性、提高其效率。政策建议主要从改善数据搜集流程、完善金融数据信息化和信息共享系统、建立不同时域和区域的预警机制三个方面展开讨论。
[Abstract]:The financial world is changing-every day there are new financial products, new financial regulations and new financial risks flowing into the financial system, and for China, where financial markets are booming, These are especially worthy of attention. The safe operation of the macro-financial system is related to the stability of the whole country and society. How to find the financial risk factors in the financial system and macroeconomic system, And how to effectively capture the macro financial system risk information and build an effective risk early warning system according to these financial risk factors can prevent the occurrence of macro financial risks, and this is the main purpose of this paper. The main idea of this paper is to construct two models to measure and predict the risk of macro-financial system-namely the construction of early-warning system. In domestic and international academic circles, many famous scholars have made outstanding contributions in this field. They have proposed and tested many models, which will be briefly reviewed one by one in the literature review at the beginning of this paper. In this paper, factor analysis is used to measure the risk of macro financial system, and BP neural network is used to predict the risk of macro financial system and build an early warning system. The two models complement each other. Complementing each other is one of the highlights of this paper. In the two models, factor analysis is relatively traditional, which can help scholars to excavate the risk factors of macro financial system more accurately and to reduce the dimension of input variables at the same time, while BP neural network is a relatively new statistical research tool. Its application in the measurement of financial system risk and the construction of early warning system is even more important in academic circles, and its excellent performance in nonlinear function fitting makes it a good choice for financial data analysis tools. This paper is a combination of the traditional model and the emerging model, but whether traditional or emerging, the results of measurement and prediction produced by the model are cogent with reality. The model is built to understand reality more clearly, but the model is only a means, not an end; therefore, based on the output of the model, The paper further expounds the economic significance of the risk factors of the macro-financial system and the macro-financial sub-system, and the practical significance and instructive effect of the measurement and prediction results. At the end of the paper, the author expounds the imperfections of the models used in the paper, and puts forward some policy suggestions to enhance the effectiveness and efficiency of the risk early warning system of the macro financial system. The policy suggestions are mainly discussed from three aspects: improving the data collection process, perfecting the financial data information and information sharing system, and establishing the early warning mechanism in different time domain and region.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832

【参考文献】