组合模型在深圳港货物吞吐量预测中的应用

Application of combined model in forecasting cargo throughput capatity of Shenzhen port

根据深圳港货物吞吐量的历史数据,分别利用时间序列的三次指数平滑法、GM(1,1)灰色预测模型法以及组合预测法对深圳港货物吞吐量进行探讨,并对未来几年深圳港货物吞吐量进行预测.结果表明:组合预测模型融合了前两种预测方法的优点,其平均绝对误差、平均相对误差的绝对值,都小于单独用三次指数平滑法或GM(1,1)灰色预测模型法时的平均绝对误差、平均相对误差的绝对值,可以降低误差,提高预测精度.

The paper analyzed the cargo throughput capatity of Shenzhen port by using cubic exponential smoothing,GM(1,1) model and combined forecasing method,and forecast the cargo throughput capatity of Shenzhen port in the next few years.The result implied that by melting the advantages of the two former forecast methods,combined forecasting model's mean absolute error,mean relative error are respec- tively,less than those when using cubic exponential smoothing or GM(1,1) model respectively,the com- bined forecasting model could reduce error and increas the precision of forecast.