基于门限自回归模型对人民币汇率波动的研究  

Research on the Fluctuation of the RMB Exchange Rate based on the Threshold Autoregressive Model

作　　者：张茗慧 潘金凤 史倩 Zhang Minghui;Pan Jinfeng;Shi Qian(School of Mathematics and Statistics,Weifang University,Weifang,261061,Shandong;School of Aviation Service and Maintenance,Ji'nan Technician College,Jinan,250031,Shandong)

机构地区：[1]潍坊学院数学与统计学院,山东潍坊261061 [2]济南市技师学院航空服务与维护系,山东济南250031

出　　处：《中国商论》2024年第23期107-112,共6页China Journal of Commerce

摘　　要：人民币汇率的波动受多方面因素的影响,为研究人民币汇率的波动情况,进一步预测人民币汇率的变化,本文以2011年1月—2023年12月的人民币兑美元的汇率中间价的月度数据作为研究对象,分别建立了求和自回归滑动平均模型和门限自回归模型,并对两种模型的预测结果进行对比。结果表明:人民币汇率的波动具有非线性特征,并且门限自回归模型比求和自回归滑动平均模型准确度更高,在未来几个月人民币兑美元中间价略有下降,但能够维持在相对稳定水平,人民币略有升值。The fl uctuation of the RMB exchange rate is affected by many aspects.In order to study the fl uctuation of the RMB exchange rate and further predict the change of the RMB exchange rate,this paper takes the monthly data of the central parity of RMB against the US dollar from January 2011 to December 2023 as the research object,establishes ARIMA model and TAR model,and compares the prediction results of the two models.The results show that the fl uctuation of the RMB exchange rate has nonlinear characteristics,and the TAR model is more accurate than the ARIMA model.In the next few months,the central parity of the RMB against the US dollar will decline slightly but remain at a relatively stable level,and the RMB will appreciate slightly.

关 键 词：求和自回归滑动平均模型 门限自回归模型 人民币汇率 股份指数 ARIMA模型 

分 类 号：F822[经济管理—财政学] F037.1