基于ARIMA和GRNN模型对人民币汇率的预测  

Forecasting RMB Exchange Rate Model Based on ARIMA and GRNN

作　　者：张艳[1] 蔡光兴[1] Zhang Yan Cai Guang-xing(College of science, Hubei university of technology,430068, Wuhan, Hubei ,Chin)

机构地区：[1]湖北工业大学理学院,湖北武汉430068

出　　处：《特区经济》2017年第2期53-55,共3页Special Zone Economy

摘　　要：为了提高人民币汇率预测精度,本文分别采用求和自回归移动平均(ARIMA)模型与广义自回归神经网络(GRNN)模型来拟合和预测人民币兑美元日汇率值,并比较两种模型对人民币汇率拟合预测的效果优劣。研究结果表明,GRNN模型预测效果优于ARIMA模型的预测效果,可应用于人民币兑美元汇率的预测。In order to improve the prediction accuracy of the RMB exchange rate, this paper presents an comparative approach to forecaste daily data on RMB exchange rate which making autoregressive integrated moving average（ARIMA） and generalized regression neural network（GRNN） models. The conclusion shows that the predicted result of GRNN model is more fitted than that of ARIMA model and GRNN model can be applied to forecasting RMB exchange rate.

关 键 词：ARIMA模型 GRNN模型 汇率预测 

分 类 号：F224.9[经济管理—国民经济]