摘要
在数据搜集中常常因为成本过高或难度过大而不能得到真正的面板数据。而组群分析方法利用不同年度的横截面数据构建出"伪面板",尽管"伪面板"数据并非真正的面板数据,但它的优势是可处理样本损失和测量误差问题。文章介绍了如何在忽略测量误差情况下推导"伪面板"的标准组内估计量一致性应满足的条件,以及在一致性条件不能满足条件下偏差和估计方差的表达式。
In data collection, real panel data cannot be obtained because of high cost or difficulty. However, the method of group analysis uses cross-sectional data of different years to construct "pseudo-panel". Although the "pseudo-panel" data is not the real panel data, its advantage is that it can be employed to deal with the problems of sample loss and measurement errors. This paper introduces how to deduce the criteria for consistency of estimators within the standard group of "pseudo-panels" to satisfy when the measurement error is ignored, and the expression of deviation and estimated variance under the condition that the consistency condition can not satisfy the condition.
作者
杨琦
Yang Qi(Economic Research Center of Upper Yangtze River,Chongqing Technology and Business University,Chongqing 400067,China;Financial Monetary Institute,Chongqing Technology and Business University,Chongqing 400067,China)
出处
《统计与决策》
CSSCI
北大核心
2019年第21期22-25,共4页
Statistics & Decision
基金
国家社会科学基金西部项目(13XJY023)
关键词
组群分析
伪面板
适用条件
group analysis
pseudo-panel
application conditions