基于关联面积法的物流货运量组合预测模型

针对物流货运量组合预测模型中赋权策略这个难点问题,在灰色模型、三次多项式趋势外推模型(PTEM)和三次指数平滑模型(TESM)基础上,引入关联面积法确定组合权系数,构建物流货运量组合预测模型。实例表明,与等权法、熵权法、平均绝对误差法对比分析可知,关联面积法综合体现了模型预测结果与真实时间序列之间的相关性及拟合误差,提高了模型预测性能和精度,是一种有效的组合赋权策略。

Adaptive combination forecasting model for logistics freight volume based on area correlation method

The forecasting performance of combined model is typically influenced by the combination weights assignment.A new combination weights assignment approach based on area correlation method was proposed.This study applied grey model,Polynomial Trend Extrapolation Model(PTEM) and Triple Exponential Smoothing Model(TESM) to develop a combination forecasting model to predict logistics freight volumes,in which the coefficients of combination weights were determined by area correlation method.The new method based on area correlation method shows its superiority in determining combination weights,compared with some other combination weight assignment methods such as equal weight method,entropy weight method and reciprocal of mean absolute percentage error weight method.Since area correlation method can comprehensively evaluate both the correlation and fitting error of forecasting model,it is an effective approach to determine the combination weights.