摘要
人民币正式加入SDR以来,人民币汇率的波动日益复杂,而汇率变动特征的识别是预测人民币汇率的关键环节。为提取出人民币汇率波动的特征规律,本文选择对2016年10月10日—2020年10月10日的人民币兑美元日序列建立模型并预测。同时,本文为评价ARIMA-GARCH组合模型的预测效果,将其与ARIMA模型预测结果相互对比。结果显示,ARIMA-GARCH模型不仅可以提取出人民币汇率波动的规律性,还可以改良ARIMA模型的预测精度,预测精度最高,在短期内能够得到较好的预测值。
Since the offi cial inclusion of the RMB into the SDR basket,the fl uctuation of the RMB exchange rate has become increasingly complex,and the identifi cation of exchange rate movement patterns is a crucial step in forecasting the RMB exchange rate.To extract the characteristic patterns of RMB exchange rate fl uctuations,this paper selects the daily RMB-to-USD series from October 10,2016,to October 10,2020,to build a model and make predictions.Meanwhile,to evaluate the predictive performance of the ARIMA-GARCH model,this paper compares its results with those of the ARIMA model.The results show that the ARIMA-GARCH model can not only extract the regularity of RMB exchange rate fl uctuations but also improve the predictive accuracy of the ARIMA model,thereby achieving the highest level of prediction precision and providing accurate forecasts in the short term.
作者
王艺柳
WANG Yiliu(Anqing Normal University,Anqing,Anhui 246011)
出处
《中国商论》
2024年第11期9-12,共4页
China Journal of Commerce
基金
安徽省高等学校科学研究重点项目(哲学社会科学)“高水平对外开放的特征、机制与效应”(2023AH050459)。
