统计研究

• 论文 • 上一篇    下一篇

高阶矩、HCCA模型与银行系统风险前瞻预判

张立华 丁建臣   

  • 出版日期:2016-01-15 发布日期:2016-01-22

Prospective Prediction of Banking Systemic Risk with HCCA Mode

  • Online:2016-01-15 Published:2016-01-22

摘要: 本文基于Corrado-Su的理论及修正,扩展了CCA模型,选取中国14家银行A股股价和资产负债表数据,测度出14家银行单个和联合违约距离及系统期望损失等风险指标。本文模型较好地反映了美国金融危机和欧洲主权债务危机,基本刻画了中国的实际经济,实证结果发现:(1)HCCA模型大致提前3-6个月或更长预测到全球金融危机和欧洲主权债务危机的来临,给金融部门防范金融危机爆发争取到了宝贵的时间;(2)HCCA模型显示中国银行业系统风险在高位不断累积,对金融监管当局和商业银行管控金融宏观风险提出了有益的警示。本文的政策建议是,应度量引入高阶矩(偏度和峰度)的隐性资产价值对违约距离和期望损失等风险指标的影响,从而提高金融政策的准确性和有效性,加强政策措施的前瞻性预判。

关键词: HCCA模型, 高阶矩, 违约距离, 系统风险

Abstract: Based on Corrado-Su's formula and correction, it extends the CCA model, and measures the risk indicators such as individual DD, systemic DD and expected loss among 14 banks with their stock price and balance sheet data. The model reflects the dynamic processes of the global financial crisis, European sovereign-debt crisis, and the China’s real economy well. It finds that: (1) HCCA model can predict global financial crisis and European sovereign-debt crisis ahead of three to six months or longer, and thus help the financial sector catch the valuable time for to prevent and manage the financial crisis; (2) HCCA model shows that China’s banking systemic risks are steadily accumulated, and puts forward the early warning of the bank’s systemic risks for the financial authorities and commercial banks. It is necessary to measure the effect of hidden capital value with higher moments on the risk indicators such as DD and expected loss. It can improve the accuracy and effectiveness of financial policies, and strengthen the forward-looking prediction of policy measurements.

Key words: HCCA Model, Higher Moments, Distance to Default, Systemic Risk