摘要
基于时间序列理论建立SARIMA和SARIMA-LSTM组合模型,以求对河南省快递业务量发展趋势的精准预测。由于传统的时序模型在快递业务量预测中很难捕捉到数据序列中的非线性特征,因此研究提出了一种季节性差分回归移动平均模型(SARIMA)与长短期记忆网络(LSTM)相结合的组合预测模型。通过对这2种模型的预测结果进行对比分析,发现SARIMA-LSTM组合模型在对快递业务量变动趋势的预测上具有更高的准确性。
In order to realize the effective prediction of the development trend of express quantity in Henan Province,based on the time series theory,the study established SARIMA and SARIMA-LSTM combination models to respectively forecast the data of express quantity in Henan Province.Since it is difficult for the traditional time series model to capture the nonlinear features in the data series in express quantity forecasting,the study proposes a combined forecasting model combining the seasonal differential regression moving average model(SARIMA)and the long and short-term memory network(LSTM).By comparing and analyzing the prediction results of these two models,it is discovered that the SARIMA-LSTM combination model has higher accuracy in predicting the trend of express quantity.
作者
张美悦
桂海霞
ZHANG Meiyue;GUI Haixia(Faculty of Economics and Management,Anhui University of Science and Technology,Huainan 232001,China)
出处
《安阳工学院学报》
2024年第3期96-103,共8页
Journal of Anyang Institute of Technology
基金
国家自然科学基金“基于agent偏好和资源约束的重叠联盟机制研究(61703005)”
安徽省高校优秀科研创新团队基金“煤矿安全与能源环境治理(2022AH010054)”
安徽省重点研发计划国际科技合作专项(202004b11020029)。
