MWTP计量模型研究与应用进展  

Research and Application Advancements of MWTP Measurement Models

作　　者：申慧敏[1] 夏赞才[1] 刘婷[1] 张悦 Shen Huimin;Xia Zancai;Liu Ting;Zhang Yue(College of Tourism,Hunan Normal University,Changsha 410081,China)

机构地区：[1]湖南师范大学旅游学院,长沙410081

出　　处：《统计与决策》2020年第23期37-41,共5页Statistics & Decision

基　　金：湖南省社会科学基金资助项目(18JD46;14YBA262);湖南省教育厅科学研究项目(18C0022)。

摘　　要：边际支付意愿(MWTP)估计的两种主要方法是离散选择实验法(DCE)和特征价格法(HPM)。文章从基本原理出发,系统性梳理、总结了各方法的经典计量模型、模型改进与拓展的建模理论及应用的最近新进展。DCE法MWTP估计模型中,潜变量Logit模型、逻辑混合Logit模型、混合选择模型和动态Logit模型是国内外近期关注的热点;Hedonic法MWTP估计模型中,改进Hedonic模型以解决案例应用中出现的样本数据缺失过多和特征价格存在空间异质性问题是研究热点。模型的设定、变量内生性、参数相关、尺度效应、模型偏差等是MWTP计量模型实际应用中备受关注的内容,建立统一的Logit模型和Hedonic模型框架进行MWTP估计是理论研究的最新动向。The two primary approaches to estimate Marginal Willingness To Pay(MWTP)are the Discrete Choice Experiment method(DCE)and Hedonic Price Method(HPM).Based on the basic principles,this paper systematically summarizes the classical measurement models and model improvements of each method and recent progress of extended modeling theory and application.In the MWTP measurement models with DCE method,the latent variable Logit model,logically mixed Logit model,hybrid choice model and dynamic Logit model are recent hot topics at home and abroad.In the MWTP measurement models with HPM,it’s a research hotspot to improve hedonic price models to deal with too much missing sample data and the spatial heterogeneity of hedonic price in the case application.The model setting,variable endogeneity,parameter correlation,scale effect and model bias are the most concerned contents in the practical application of MWTP measurement models,and the establishment of unified Logit model and hedonic model frameworks for MWTP estimation is the latest trend of theoretical researches.

关 键 词：边际支付意愿 HEDONIC模型 LOGIT模型 研究进展 

分 类 号：F064.1[经济管理—政治经济学]