基于HP滤波和ARMA-GARCH模型的人民币汇率趋势预测  

Trend Prediction of RMB Exchange Rate Based on the HP Filter and ARMA-GARCH Model

作　　者：宋博[1] 陈万义[1] 

机构地区：[1]南开大学计算机与控制工程学院,天津300350

出　　处：《数学的实践与认识》2017年第1期70-78,共9页Mathematics in Practice and Theory

摘　　要：在复杂多变的金融市场,人民币汇率的变化受多种因素的影响,而人民币汇率的变化又影响着经济生活的方方面面,人民币汇率及其变化特征受到人们的广泛关注,研究人民币汇率变化特征,正确分析与预测人民币汇率的走势,对于国家和各个经济主体制定金融政策和投资决策具有十分重要的意义,采用HP滤波法将汇率数据序列分解为趋势成分序列和波动成分序列,然后使用自回归和ARMA-GARCH模型分别进行拟合和预测,通过实证分析发现模型有着较好的预测效果,可以为金融产品的预测研究和制定金融政策提供参考。The change of RMB exchange rate has become an important variable in financial market while it is also influenced by different factors.As the variation characteristics of RMB exchange rate attract more attention,studies of variation characteristics and prediction for RMB exchange rate trend become more important in financial policy-making and investment decision-making area.In this paper,we begin with HP filter model to decompose exchange rate data series into the trend and fluctuation series,then using AR model and ARMAGARCH model to fit and forecast the data.The result of the empirical analysis presents better forecasting effect,which could strongly support the prediction research of financial products and financial policy-making.

关 键 词：HP滤波 ARMA-GARCH模型 人民币汇率趋势预测 

分 类 号：F224[经济管理—国民经济] F832.6