时间序列分析在我国农业总产值预测中的应用  

Application of Time Series Analysis in Prediction of Agricultural Output Value in China

作　　者：张镪 马元魁[1] 张建生[1] ZHANG Qiang;MA Yuankui;ZHANG Jiansheng(School of Science,Xi'an Technological University,Xi'an,Shaanxi 710021,China)

机构地区：[1]西安工业大学理学院

出　　处：《数学建模及其应用》2019年第3期13-21,共9页Mathematical Modeling and Its Applications

基　　金：国家自然科学基金项目(61807026);陕西省教育厅专项科研计划项目(2010JK585);西安市未央区科技计划项目(201843);西安工业大学校长基金项目(XAGDXJJ17027)

摘　　要：应用时间序列分析的方法,运用R软件,对我国1978-2015年各年的农业总产值数据进行了分析.首先,绘制时间序列图,计算自相关系数与偏自相关系数,根据自相关系数图与偏自相关系数图进行纯随机性检验,并在此基础上,建立了两个模型:ARIMA(1,1,3)模型和ARIMA(1,2,1)模型;然后,对这两个模型获取残差序列,对我国2016-2022年农业总产值进行了预测;最后,通过与我国2016年和2017年实际的农业总产值进行比较,结合预测模型的AIC值和残差序列的p值等多重因素考虑,认定ARIMA(1,2,1)模型拟合程度较好,可用于对我国农业总产值进行有效预测.In this paper,using the method of time series analysis and R software.the data of China's agricultural gross output from 1978 to 2015 are analyzed.Firstly,by drawing time series diagrams,calculating autocorrelation coefficient and partial autocorrelation coefficient,and testing pure randomness based on autocorrelation coefficient diagrams and partial autocorrelation coefficient diagrams,two models:ARIMA(1,1,3)model and ARIMA(1,2,1)model are established;Then the residual sequences of the two models are obtained and the gross agricultural output value of China from 2016 to 2022 is predicted;Finally,by comparing with the actual agricultural gross output value in 2016 and 2017,considering the multiple factors such as AIC value of the prediction model and p value of residual sequence,it is concluded that ARIMA(1,2,1)model fits well and can be used to effectively predict the agricultural gross output value of China.

关 键 词：时间序列分析 农业总产值 R软件 预测 

分 类 号：O213[理学—概率论与数理统计]