摘要
针对时间序列预测与曲线拟合预测的优缺点,提出了一种季节分解与曲线拟合相结合的混合预测模型,解决了曲线拟合精度与高阶次震荡性强的矛盾,提高预测的精度和模型对数据的适应能力。同时针对具有时间周期特性的物流量,该模型在预测时充分考虑了上期物流量及近期预测物流量对本期的影响。最后将该模型与其他预测方法相比较,证明该模型的有效性和可行性。
In view of the advantages and disadvantages of time series prediction and curve fitting prediction,a hybrid prediction model combining seasonal decomposition and curve fitting is proposed,which solves the contradiction between curve fitting accuracy and high-order oscillation,and improves the prediction accuracy and the adaptability of the model to data.At the same time,for the logistics volume with time-periodic characteristics,the model takes full account of the impact of the previous logistics volume and the recent forecast logistics volume on the current period.Finally,the model is compared with other prediction methods to prove the validity and feasibility of the model.
作者
晋民杰
王博
李雯
JIN Min-Jie;WANG Bo;LI Wen(Department of Transportation and Logistics,Taiyuan University of Science and Technology,Taiyuan 030024,China)
出处
《太原科技大学学报》
2021年第2期138-142,共5页
Journal of Taiyuan University of Science and Technology
基金
山西省自然基金(201701D121071)
山西省重点研发计划(201803D31076)
山西省高等学校改革创新项目(J2018128)。
关键词
物流量预测
季节分解
曲线拟合
自变量偏差
logistics volume forecast
seasonal decomposition
curve fitting
independent variable deviation