首页|中国二手车在线司法拍卖的非参数估计

中国二手车在线司法拍卖的非参数估计

扫码查看
近些年来,中国在线司法拍卖发展迅速,对该领域的实证研究具有十分重要的现实意义.这篇文章采用阿里巴巴平台上中国二手车在线司法拍卖的数据,对选择性进入模式下增价拍卖的进入成本进行了非参数估计.基于Athey和Haile(2002)以及Levin和Smith(1994)提出的识别策略,我们提出了一种两步估计法.通过对实证结果进行分析,我们得出以下四点结论:1)私人价值的经验分布符合随机占优性质,且进入后价值分布估计值的置信区间较为紧密.2)进入成本的估计值对于评估价格具有单调性,评估价格越高,进入成本估计值也会越高.3)增价拍卖获得的收益高于一价密封拍卖,且在低中高百分位数上都具有明显优势.4)当进入成本减少时,收益会有所改善;当进入成本增加时,收益会出现下降.且进入成本降低带来的收益改善幅度,比进入成本提高带来的收益降低幅度更大.
Nonparametric estimation of online judicial auctions of used cars in China
In recent years,Chinese online judicial auctions have developed rapidly,and it is of great importance to hold empirical research in this field.This paper exploits the data for Chinese online judicial auctions of used cars in Alibaba to study the entry cost in ascending auctions with selective entry.We propose a two-step estimation method based on the identification strategy of Athey and Haile(2002)and Levin and Smith(1994).From the empirical study,we have four findings:(ⅰ)The CDFs of post entry value distribution comply with the law of ordered dominance,and the confidence bands of post entry value distribution are quite tight;(ⅱ)the estimates entry cost is monotonic to the appraisal value:The higher the appraisal value,the higher the entry cost;(ⅲ)there will be a higher revenue in the ascending auction than the first-price sealed bid auction,and the advantages are significant in low,medium,and high percentiles;(ⅳ)when the entry cost is cut by 25%or 50%,the revenue is increased significantly;when the entry cost is add by 75%or 100%,the revenue is decreased significantly.Furthermore,the improvement in revenue brought by lower entry cost is greater than the reduction in revenue brought by higher entry cost.

online judicial auctionsselective entryentry costcounterfactual analysis

刘年青、许可欣

展开 >

厦门大学邹至庄经济研究院,厦门 361005

厦门大学经济学院,厦门 361005

上海财经大学经济学院,上海 200000

中国计量大学经济与管理学院,杭州 310018

展开 >

在线司法拍卖 选择性进入 进入成 反事实分析

国家自然科学基金重大项目国家自然科学基金重点项目国家自然科学基金面上项目

723943927183300471873080

2024

系统工程理论与实践
中国系统工程学会

系统工程理论与实践

CSTPCDCSSCI北大核心
影响因子:1.575
ISSN:1000-6788
年,卷(期):2024.44(9)