摘要
依据2008年陕西统计年鉴与1952～2007年陕西省GDP相关数据,采用SPSS统计软件及时间序列分析法,通过时间序列的平稳性检验、根据时间序列模型的识别规则进行定阶、模型检验、模型预测4大步骤在AIC准则下建立了ARIMA(1,2,1)时间序列模型,并根据ACF图和PACF图对模型做了适应性检验,然后对2002～2007年的实际值与预测值作比较,并利用该模型对陕西省未来6年的GDP做出预测。结果表明,各年实际值与预测值之间的相对误差均控制在5%以内,该模型的预测效果相对较好;根据模型预测的2008～2013年陕西GDP数据分别为6477.50亿、7656.62亿、9058.66亿、10735.10亿、12744.69亿、15158.20亿元,从预测结果看,陕西省的GDP在2008～2013年6年内仍将呈现出较高的增长趋势;该模型得出的预测结果只是一个预测值,而国民经济是一个复杂多变的动态系统,应随时注意经济运行中蕴藏着调整的风险,适时根据实际情况调整相应的目标值。
Based on the 2008 Shaanxi Statistical Yearbook and the relevant data of Shaanxi GDP in the years 1952-2007,SPSS statistical software and time series analysis are used to establish ARIMA(1,2,1) time series model,according to the four steps,recognition rules and stationary test of time series under AIC criterion.ACF graph and PACF graph are used to conduct the applicability test on model.Then,the actual value and predicted value in the years 2002-2007 are compared in order to forecast the GDP of Shaanxi Province in the next six years based on this model.Result shows that the relative error of actual value and predicted value is within the range of 5%,and the forecasting effect of this model is relatively good.It is forecasted that the GDP of Shaanxi Province is 647.750,765.662,905.866,1 073.510,1 274.469 and 1 515.820 billion Yuan in the year from 2008 to 2013,respectively.According to the result,GDP of Shaanxi Province shows a higher growth trend in the years 2008-2013.The forecasting result of this model is only a predicted value.But the national economy is a complex and dynamic system.We should pay attention to the risk of the adjustment in economic operation and adjust the corresponding target value according to the actual situation.
