摘要
自2005年人民币汇率改革以来,汇率波动对我国经济发展产生了很大影响,因此对人民币汇率进行研究和预测有着重要意义。鉴于金融时间序列数据的波动性和非线性特征,以及小波分析在时间序列数据分析方面的优势,运用小波分析和BP-GARCH模型相结合方法,以2014年1月2日至2017年6月30日的日人民币兑美元汇率数据为样本,实现汇率的有效预测。结果表明,该方法比BP神经网络的预测效果更佳,证明基于小波分析和BPGARCH相结合的预测模型可提高人民币兑美元汇率的预测结果。
Since the RMB exchange rate reform in2005,the exchange rate fluctuations have had a significant impact on China’s economic development.Therefore,it is of great significance to study and forecast the RMB exchange rate.In view of the volatility and nonlinear characteristics of financial time series data and the advantages of wavelet analysis in time series data,in this paper,a combination of wavelet analysis and BP-GARCH model is used to take the RMB exchange rate data from January2,2014to June30,2017as a sample to achieve effective prediction of exchange rate.The results show that the prediction effect of this method is better than that of BP neural network.It is proved that the prediction model based on wavelet analysis combined with BP-GARCH can improve the prediction effect of RMB exchange rate against the US dollar.
作者
衡亚亚
沐年国
HENG Ya-ya;MU Nian-guo(School of Management, University of Shanghai for Science and Technology, Shanghai 200093,China)
出处
《软件导刊》
2018年第12期146-150,共5页
Software Guide
