安徽财经大学安徽蚌埠 233030
对于我国外汇规模的预测，选取我国1965-2018年外汇储备数据进行分析。首先，对时间序列数据进行平稳性检验和白噪声检验，通过二阶差分消除长期趋势的影响。其次，根据序列相关图和自相关系数识别拟合AR(2)模型。再次，对AR(2)模型和参数分别进行显著性检验，重新拟合零均值的AR(2)模型。最后，进行单位根检验后对我国外汇储备规模进行预测，结果表明未来五年外汇储备余额将保持持续下降的趋势。
教育部人文社会科学研究项目“低生育水平下的中国家庭政策发展研究”（19YJCZH069）；安徽省教研项目“大数据背景下学科竞赛对新经管人才创新能力培养研究”（2018jyxm1305）。
Anhui University of Finance and economics, Bengbu,Anhui,233030,China
for the prediction of China""s foreign exchange scale, the data of China""s foreign exchange reserves from 1965 to 2018 are selected for analysis. Firstly, the stationary test and white noise test are carried out for time series data, and the influence of long-term trend is eliminated by second-order difference. Secondly, AR (2) model is identified and fitted according to sequence correlation graph and autocorrelation coefficient. Thirdly, we test the significance of AR (2) model and parameters respectively, and re fit the zero mean AR (2) model. Finally, after the unit root test, we predict the scale of China""s foreign exchange reserves, and the results show that the balance of foreign exchange reserves will continue to decline in the next five years.
