根据农产品价格变动非线性、非平稳的特点,选择SARIMA模型,以2005年1月至2010年12月大连黄瓜月度零售价格为样本,对2011年大连黄瓜短期价格变动进行预测。预测结果表明,SARIMA模型预测精度较高,平均绝对百分比误差(MAPE)为6.94%;且精度较为稳定,相对误差的标准差为4.59%。 According to the non-linear and non-stationary characteristics of the price of agricultural products, the SARIMA model was selected. Based on the monthly retail prices of cucumbers in Dalian from January 2005 to December 2010, the short-term price changes of Dalian cucumber in 2011 were predicted. The prediction results show that the prediction accuracy of SARIMA model is high, the average absolute percentage error (MAPE) is 6.94%, and the accuracy is relatively stable. The standard deviation of relative error is 4.59%.