基于GABP神经网络汇率波动弹性空间测度  

Measurement of Elastic Space of RMB Exchange Rate Volatility Based on GABP Neural Network

作　　者：黄志刚[1] 郑国忠[2] 

机构地区：[1]福州大学管理学院,福建福州350002 [2]厦门大学经济学院,福建厦门361005

出　　处：《当代财经》2013年第9期49-60,共12页Contemporary Finance and Economics

基　　金：国家自然科学基金项目"国际金融危机背景下的汇率制度风险控制及汇率弹性空间研究"(70973021);福建省社科基金重点项目"基于我国汇率制度风险控制的汇率弹性空间研究"(2009A032)

摘　　要：汇率波动弹性空间是有管理的浮动汇率制或中间汇率制的管理核心,因此,如何确定或测度汇率波动弹性空间就成为科学管理汇率波动的关键问题。实证结果表明:实际汇率波动指标(ESERV)是测度汇率波动弹性空间的良好选择;基于GABP神经网络模型的测度方法也要优于基于调整型EWMA模型方法。应用新测度方法的实证结果还表明,截止2012年1月,人民币汇率波动弹性空间已经变为1%;其政策含义是,自2012年1月底始,人民币汇率日均波幅0.5%的限制应放宽至1%。Elastic space of exchange rate volatility is the core of management under the managed floating exchange rate system or the intermediate exchange rate system. Thus, how to determine or measure the elastic space of exchange rate volatility has become the key to the scientific management of exchange rate volatility. The empirical results show that the effective exchange rate volatility index （ESERV） is a fine choice for measurement of elastic space of exchange rate volatility; the measuring method based on the GABP neural network model is also better than the method based on the regu- lative EWMA model. The empirical results with the new measuring method also indicate that up to January of 2012 the elastic space of RMB exchange rate volatility has turned into 1%. Its political implication is, starting from the end of January of 2012, the limit of average daily 0.5% amplitude on RMB exchange rate should be relaxed to 1%.

关 键 词：汇率波动弹性空间 风险价值 GABP神经网络 

分 类 号：F830.92[经济管理—金融学]