摘要
针对港口物流预测中GM(1,1)模型在预测上存在的局限性,为了进一步提高预测的准确性,本文基于样本数据序列级比和可容覆盖范围,提出了对模型数据进行线性平移的一次修正模型,然后根据一次修正模型的残差序列建立BP残差二次修正模型,并应用GM(1,1)模型和修正后的模型分别对青岛港货物吞吐量进行了拟合和预测。通过实例验证,可以看到拟合值的平均相对误差由10.34%降低到0.37%,预测值的平均相对误差由27.9%降低到1.32%。结果表明,GM(1,1)修正模型在预测上具有更高的准确性,可以有效预测港口的货物吞吐量。
In view of the limitations of GM(1,1)model in port logistics prediction,this paper proposes to modify GM(1,1)model twice to further improve the accuracy of prediction.Firstly,based on the sequence level ratio and tolerable coverage of sample data,A modified model for linear translation of model data is proposed.Then the BP residual secondary correction model is established according to the residual sequence of the primary correction model.The GM(1,1)model and the modified model are used to fit and predict the cargo throughput of Qingdao port.Through the example verification,it can be seen that the average relative error of the fitted value is reduced from 10.34%to 0.37%,and the average relative error of the predicted value is reduced from 27.9% to 1.32%.The results show that the GM(1,1)modified model has higher accuracy in prediction and can effectively predict the cargo throughput of the port.
作者
陈慧敏
孙晓燕
段晓亮
陈虹
Chen Huimin;Sun Xiaoyan;Duan Xiaoliang;Chen Hong(Beijing Institute of Economics and Management,Beijing 100102;Guangxi University of Finance and economics,Nanning 530003)
出处
《北方经贸》
2022年第3期23-27,共5页
Northern Economy and Trade
基金
北京经济管理职业学院校级课题(19YBA31)
北京经济管理职业学院博士资助课题(20BSA03)。
