基于潜在类别模型的高铁旅客市场细分

发布时间：2019-04-17 08:03

【摘要】：对高速铁路旅客市场进行细分是应用收益管理理论的重要环节.基于京沪高铁的客票数据,选取年龄、性别、出行日期、出行距离、购票方式和提前购票时间6类外显变量作为分类指标,采用潜在类别模型进行高速铁路旅客市场的细分.首先将外显变量概率参数化后代入模型进行建模并利用Mplus软件进行模型求解,模型拟合的AIC和BIC等指标表明,当潜在类别为3类时模型具有较好的效果.然后根据模型参数估计结果对所有数据进行潜在聚类分析,分类正确率达到93%左右,表明分类结果合理,3种类别的旅客在提前购票时间、出行距离等方面具有明显的差异.潜在类别模型的引入可以为我国高速铁路收益管理理论研究和实践应用提供参考借鉴.
[Abstract]:Subdivision of high-speed railway passenger market is an important link in applying revenue management theory. Based on the ticket data of Beijing-Shanghai high-speed railway, six explicit variables, such as age, gender, travel date, travel distance, purchase mode and advance purchase time, are selected as classification indexes, and the potential category model is used to segment the passenger market of high-speed railway. First, the probability parameterized offspring of explicit variables are built into the model and solved by Mplus software. The AIC and BIC indexes of model fitting show that the model has a good effect when the potential class is three classes. Then all the data are analyzed by potential cluster analysis according to the result of model parameter estimation, and the classification accuracy is about 93%, which shows that the classification result is reasonable, the passengers of the three categories have obvious differences in the advance purchase time, travel distance and so on. The introduction of potential category model can provide reference for the theoretical research and practical application of high-speed railway revenue management in China.
【作者单位】：北京交通大学交通运输学院;
【基金】：国家自然科学基金项目(u1434207) 中国铁路总公司科技研究计划项目(2015X006-G)~~
【分类号】：F532.3

【参考文献】