我国物流景气指数(LPI)预测模型的构建及分析  

Construction and Analysis of China Logistics Prosperity Index(LPI) Forecasting Model

作　　者：朱东红 钱叶霞 陈东清 Zhu Donghong;Qian Yexia;Chen Dongqing(School of Economics & Management,Fuzhou University,Fuzhou 350000,China)

机构地区：[1]福州大学经济与管理学院,福建福州350000

出　　处：《物流技术》2019年第1期88-93,共6页Logistics Technology

基　　金：福建省本科高校专业(物流管理)综合改革试点项目(160418)

摘　　要：物流景气指数反映了一个国家物流业发展运行的总体情况,是衡量物流业活跃程度的关键指标,它与货运量、港口货物吞吐量、快递业务量等众多物流相关指标有着密切联系。对于物流景气指数的预测分析,有助于一国宏观经济政策的制定和实施。以物流景气指数为对象,使用GM(1,1)预测、BP神经网络预测和ARIMA预测三种模型对物流景气指数进行建模分析,通过三种模型的误差比较,结果发现ARIMA模型预测拟合度最佳,LPI的预测值通过二次差分转换后几乎落入到95%的置信区间内,真实值和预测值吻合度最高,很好地拟合了物流景气指数在时间序列上的变化趋势,最后以该模型预测分析2018年8月到12月的LPI数值。The logistics prosperity index could reflect the overall development and condition of the logistics industry of a country.It is a key indicator to measure the activity level of the logistics industry and is closely related to many logistics indicators such as freight volume, port cargo throughput and express delivery volume.The forecasting and analysis of the logistics prosperity index would help the formulation and implementation of the country's macroeconomic policies.In this paper,the logistics prosperity index is modeled and analyzed using GM (1,1),BP neural network and ARIMA.Through the error comparison of the three,it is found that the ARIMA model has the best prediction-to- reality fitness.At the end,this model is used to predict and analyze the LPI value of China between August and December 2018.

关 键 词：物流景气指数 GM(1 1)预测 BP神经网络预测 ARIMA模型预测 建模分析 

分 类 号：F224.0[经济管理—国民经济] F252