基于支持向量回归机的区域物流需求预测模型及其应用*

为了提高区域物流需求预测的能力，从区域经济等影响因素指标与区域物流需求之间的内在关系的角度,应用基于结构风险最小化准则的支持向量回归机（SVR）方法, 建立“影响因素—区域物流需求” SVR预测模型来研究预测区域物流需求问题。在选择适当的参数和核函数的基础上，对上海市物流需求量进行预测，发现该方法能获得较小的训练相对误差和测试相对误差。

Model of regional logistics demand forecasts based on support vector regression and its application

To improve the forecast ability of regional logistics demand,from the aspect of the intrinsic relations,between the factors of the impact(regional economic and so on)and regional logistics demand,support vector regression(SVR) based on structural risk minimization is applied,by establishing the factors of impact-regional logistics demand SVR forecast model to forecasting regional logistics demand.By selecting appropriate parameters and kernel function,the proposed approach is used for forecasting logistics demand of Shanghai city,experimental results show that the above method obtained less training relative error and testing relative error is lesser.