fixed effects models are biased ?

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

我们如何理解这句话：
“most fixed-effects models produce biased and inconsistent parameter estimates (Bernard and Jensen, 2004)”. Why? 既然这样这么还有这么多应用呢？

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：fixed effect effects models Effect Biased models Fixed effects Biased

这个论断应该是有前提的吧。
经查，发现楼主有点断章取义：Bernard, A., J. Jensen, 2004, Why some firms export, Review of Economics and Statistics, 86 (2): 561-569. 的原文如下：

There are several potential estimation strategies for this dynamic binary
choice framework with unobserved heterogeneity, including probit with random
or fixed effects, conditional logit, and linear probability models with
fixed or random effects. A starting point in choosing among the available
specifications is the decision whether unobserved plant heterogeneity is better
modelled as fixed or random effects. The use of random effects requires
that the plant effects be uncorrelated with the regressors. Most fixed effects
models, on the other hand, produce biased and inconsistent parameter
estimates, especially for the coefficient on the lagged dependent variable.

他们谈论的是动态 logit 或 probit 模型的估计方法，FE当然是有偏的。

谢谢ARLION的指点，小弟明白了 下次再认真些。。。