快递需求量组合预测模型构建及实证研究

发布时间：2018-08-28 12:10

【摘要】：随着世界贸易和国内贸易活动变得更加活泼,快递在社会经济活动中扮演着越来越重要的角色,带动了其他经济产业的发展,因此受到政府的高度重视。政府相继出台快递业指导意见和规划,引导我国快递业稳步发展。快递需求量预测是快递行业规划的基础,鉴于此,本文把快递需求量预测作为研究对象,旨在建立合适的快递需求量预测模型,具有一定的实用价值。本文根据现有快递业研究相关文献,联系我国实际情况,分析了我国快递需求量的特点、快递需求量的影响因素以及快递需求量预测步骤。考虑数据可获得性,选取了 GDP、社会消费品零售总额、邮政业务量、网民人数以及货物周转量等与快递需求量相关的8个影响指标,构建预测指标体系。通过灰色关联法定量分析了我国快递需求量与这8个指标之间的灰色关联度,得出从2006年至今,我国快递需求量受到进出口总额和邮政业务量的影响最大的结论。为了达到便于操作、准确度高、适用性强的目的,从现有的预测方法中选出时间序列模型中的灰色预测模型、趋势外推法和因果关系模型中的多元线性回归模型,结合Shapley值分配原理,建立组合预测模型。最后以四川省快递需求量为研究对象,搜集2006-2016年四川省相关自变量统计指标,对四川省快递需求量进行预测和验证,结果表明,组合预测模型适用于快递需求量预测,准确度较高,满足实际需求,但是只能用于中短期预测;四川省快递需求量在未来几年仍然继续增长。
[Abstract]:With the world trade and domestic trade activities becoming more and more lively, express delivery plays an increasingly important role in social and economic activities, driving the development of other economic industries, so the government attaches great importance to it. The government has issued express industry guidance and planning to guide the steady development of China's express industry. Express demand forecast is the basis of express delivery industry planning. In view of this, this paper regards express demand forecast as research object, aiming at establishing suitable express demand forecasting model, which has certain practical value. According to the related documents of express delivery industry and the actual situation of our country, this paper analyzes the characteristics of express demand in China, the influencing factors of express demand and the forecasting steps of express demand. Considering the availability of data, this paper selects 8 indexes related to the demand of express delivery, such as the total retail volume of consumer goods in GDP, postal business volume, the number of Internet users and the volume of goods turnover, and constructs the prediction index system. The grey correlation degree between the demand for express delivery and the eight indexes is quantitatively analyzed by the grey correlation method, and the conclusion is drawn that the demand for express delivery in China has been most affected by the total import and export volume and the volume of postal business since 2006. In order to achieve the purpose of easy operation, high accuracy and strong applicability, the grey prediction model, the trend extrapolation model and the multivariate linear regression model in the time series model, the trend extrapolation method and the causality model are selected from the existing prediction methods. Combined with the principle of Shapley value allocation, the combined prediction model is established. Finally, take Sichuan express demand as the research object, collect the relevant independent variables statistical index of Sichuan province from 2006 to 2016, forecast and verify the Sichuan express demand. The result shows that the combined forecasting model is suitable for express delivery demand forecast. Accuracy is high, meet actual demand, but can only be used in the short and medium term forecast; Sichuan express demand will continue to grow in the next few years.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F259.2

【参考文献】