中美汇率非线性波动特征及预测

发布时间：2018-02-23 21:23

本文关键词： 非线性波动人民币汇率结构转换平滑移动自回归异方差模型　出处：《统计与决策》2012年22期 　论文类型：期刊论文

【摘要】：文章选取1994年1月至2011年2月美元兑人民币月平均汇率作为研究对象,首先建立STAR模型,分析了中美汇率的非线性特征,通过检验发现ESTAR模型能较好拟合汇率增量序列;其次对ESTAR-ARCH类三种组合模型进行比较,结果表明ESTAR-GARCH模型能更好地描述汇率变化的趋势,说明汇率增量既具有对称的非线性特征,也存在波动的群集效应;最后基于ESTAR-GARCH(1,1)模型对近两年来美元兑人民币月平均汇率进行预测,预测精度很高,说明所建模型非常合理,并且预测结果显示未来十个月人民币将继续升值。
[Abstract]:From January 1994 to February 2011, the average monthly exchange rate between US dollar and RMB is selected as the research object. Firstly, the STAR model is established to analyze the nonlinear characteristics of the exchange rate between China and the United States. It is found that the ESTAR model can fit the increment sequence of exchange rate. The results show that the ESTAR-GARCH model can better describe the trend of exchange rate change, indicating that exchange rate increment has both symmetric nonlinear characteristics and fluctuation cluster effect. Finally, based on the ESTAR-GARCH1) model, the average monthly exchange rate of the US dollar against the RMB in the past two years is forecasted. The accuracy of the prediction is very high, which shows that the model is very reasonable, and the forecast result shows that the RMB will continue to appreciate in the next ten months.
【作者单位】：武汉理工大学理学院;武汉理工大学经济学院;
【基金】：国家社会科学基金资助项目(09BJY022) 中国博士后基金资助项目(20100471168)
【分类号】：F224;F832.6

【参考文献】