我国影子银行风险预警模型的建立与实证研究

An Empirical Study on Construction of Risk Early Warning Model of Shadow Banking in China

通过构建BP神经网络模型,本文分析了我国影子银行体系风险状况,并进行风险等级评价。实证结果表明:2005-2009年影子银行体系的风险增加比较明显,2009-2013年的风险增加虽然比较缓慢,但安全形势仍然日趋严峻;社会融资规模的急剧增大对影子银行的风险有非常明显的促进作用;影子银行风险变化规律与股市行情的变化趋势非常相近,间接证实了风险传导链的存在;BP神经网络模型训练后的期望输出和实际输出基本符合,误差在可接受范围之内,说明训练后的BP神经网络达到精度要求,可以对我国影子银行体系的安全状况进行预警。

In this paper,the risk status and level of China＇s shadow banking system were analyzed by a BP neural network model. The results showed that the risk of shadow banking system increased obviously from 2005 to 2009,and increased slowly from 2009 to 2013,but the security situation was still stern; the financing from nongovernment sources had a very obvious promoting effect on risk of shadow banking; the change trends of stock market were very obvious and similar to the risk of China＇ s shadow banking in China,which indirectly confirmed the existence of risk transmission chain; the desired output of the trained BP neural network model accorded with the actual output,and the test output was very close to the expected value,which could early warn the safety status of China＇s shadow banking system.