极端金融事件对系统性风险的影响分析——以中国银行部门为例
阅读全文
下载全文 |
| Title | An Analysis of the Impacts of Extreme Financial Events on Systemic Risk——Evidence from China's Banking Sector
|
| 作者 | 唐文进 苏帆 |
| Author | Tang Wenjin and Su Fan |
| 作者单位 | 中南财经政法大学 |
| Organization | School of Finance, Zhongnan University of Economics and Law |
| 作者Email | wjtang@aliyun.com；amamiyas@163.com |
| 中文关键词 | 系统性风险；极端金融事件；宏观跳跃未定权益分析；混频宏观动态因子 |
| Key Words | Systemic Risk; Extreme Financial Events; Macro-Jump-CCA; Mixed- Frequency Macro Dynamic Factor |
| 内容提要 | 极端金融事件往往因为发生概率极小而被主流研究所忽略，而且，基于平稳随机过程的传统理论和模型难以刻画和分析风险突然全面爆发的现象。本文的边际贡献在于，从理论上将风险激增的非线性机制纳入考量，改进已有模型使之更贴近极端金融事件的实际；从方法上探索利用改进后模型预警系统性风险的可行性并验证其先进性。以中国银行部门为例的研究结果表明，本文提出的跳跃未定权益分析模型将传统模型的连续扩散假设放松为跳跃扩散假设，能更好地刻画极端金融事件的风险激增特征，并可比传统研究提前大约3~6个月预警其系统性风险；而如果进一步纳入本文创造性地综合利用金融市场和宏观经济信息构建的混频宏观动态因子作为风险信息源，就还可提前识别金融市场噪声信号并降低其影响，即使在噪声条件下也能为防范系统性风险提供2~3个月的政策反应时间。 |
| Abstract | Extreme financial events are ignored by mainstream researches because they are usually taken as extremely low probability events or "black swan" events, and furthermore, it's quite difficult for traditional theories and models based on stationary stochastic process hypothesis to describe and analyze the phenomenon of sudden outbreak of risk. Our marginal contributions are as follows: (1) theoretically bringing the nonlinear mechanism of risk surge into consideration and improving the existing model to make it more close to the reality of extreme financial events; (2) methodologically exploring the feasibility of using the improved model to early warning systemic risk and verifying its advancements. The empirical results based on China's banking sector show that the paper replaces pure diffusion hypothesis in conventional contingent claims analyses (CCA) with jump diffusion hypothesis, thus better catch the risk surge characteristics of extreme financial events, and can detect systemic risk about 3~6 months ahead of traditional studies. Considering that the jump risk caused by extreme financial events is reflected not only in financial markets, the paper constructs macroeconomic dynamic factors based on information from both financial markets and macro-economy to improve forward reflective effects, hence a Macro-Jump-CCA model which is more applicable to early warning of systemic risk. Through our empirical analyses based on China's banking sector, we find that Macro-Jump-CCA can identify noise signals and reduce adverse effects, and therefore can provide 2~3 months for policy reaction to precautions against systemic risk even in condition of noise signals. |
| 文章编号 | WP1194 |
| 登载时间 | 2017-06-16 |
|