摘要
本文介绍了符合金融系统预测规律的ARIMA时间序列模型,并根据我国货币供应量实际数据对2008年5月-2009年4月货币供应量走势进行了预测检验。实证预测结果显示与实际№相对照,模型预测精度较高,平均相对误差绝对值仅为1.56%,说明ARIMA模型能比较准确地预测我国货币供应量走势,可为我国货币供应量的预测和走势提供可靠的参考依据,并由此预计在2009年9月货币供应量将突破60万亿元。
In this article, we introduce the ARIMA time series model, which confirms to the law of the financial system in line with forecast. Based on the actual data of money supply in our country,we carry on a forecast test on money supply trend between May 2008 - April 2009 and the forecasting result is consistent in the actual M2. The accuracy of the model prediction is high,and the mean absolute percentage error is only 1.56 percent, which illustrates the ARIMA model can predict the trend of money supply with high accuracy, that it can provide the reliable result on the prediction and the trend of money supply in our country. According to the ARIMA model, money supply in our country will exceed 60 trillion yuan till September, 2009.
出处
《数学理论与应用》
2009年第4期69-74,共6页
Mathematical Theory and Applications