基于R语言的我国甘蔗产量ARIMA模型建立与预测分析

Establishment of ARIMA Model and Forecast for Sugar Cane Production in China Based on R Language

采用时间序列预测法对我国甘蔗产量预测问题进行了研究。以国家统计局1965~2012年间我国历年甘蔗产量统计数据为基础,依据自回归积分滑动平均（ARIMA）模型理论,运用R语言建立了ARIMA模型对我国未来甘蔗产量进行了预测。数据检验显示：该模型拟合效果较好,预测精度较高,表明运用该模型对我国甘蔗产量的变化趋势进行了分析及预测是可行的,具有较高的实际应用价值。分析了该模型所得预测结果,对我国甘蔗产业未来发展的可预见性趋势与风险提供了应对建议。

This paper studied the prediction of the yield of sugar cane in China with time series forecasting mettloci. Based on statistical data of sugar cane annual production during 1965--2012 years provided by National Bureau of Sta- tistics, this paper established an autoregressive integrated moving average （ARIMA） model with R Language to pre- dict the future of sugar cane production in China, on the basis of ARIMA model theory. Data test showed that the model fitted the data well, with good precision. The test result also indicated that utilizing the model to forecast and analyze the trend for sugar cane production of China is feasible, with a high practical value. Through the analysis of the prediction result of the model, this paper provided advice on trend forecasting and risk response for the future de- velopment of sugar cane industry of China.