stata计量高手请进

mzdg

3053

收藏 2017-03-23

面板数据用xtreg y x, fe和areg y x, absorb (id)做出来的系数完全一致，但是两者的聚类稳健标准误存在较大差异，请问这是什么原因？

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

全部回复

黃河泉

2017-3-23 16:10:53

请参考 http://www.statalist.org/forums/ ... which-one-is-better 与 http://www.statalist.org/forums/ ... be-more-appropriate。

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

mzdg

2017-3-23 17:31:31

黃河泉发表于 2017-3-23 16:10
请参考 http://www.statalist.org/forums/forum/general-stata-discussion/general/5807-areg-and-xtreg-fe ...

非常感谢

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

黃河泉

2017-3-23 17:40:58

mzdg 发表于 2017-3-23 17:31
非常感谢

You are welcome.

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

蓝色

2017-3-23 18:30:46

黃河泉发表于 2017-3-23 16:10
请参考 http://www.statalist.org/forums/forum/general-stata-discussion/general/5807-areg-and-xtreg-fe ...

xtreg命令的 dfadj 选项在stata 9 的help中有介绍，但是stata10 以后就没有介绍了。

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

黃河泉

2017-3-23 18:38:59

蓝色发表于 2017-3-23 18:30
xtreg命令的 dfadj 选项在stata 9 的help中有介绍，但是stata10 以后就没有介绍了。

老实说，我从来不知道有这"玩意"，哈哈！

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

点击查看更多内容…

蓝色

2017-3-23 21:07:39

黃河泉发表于 2017-3-23 18:38
老实说，我从来不知道有这"玩意"，哈哈！

我也是刚查的。
以前版本的stata有许多选项都很好用，但新版的都不介绍了

help xtreg                                                                               dialog:  xtreg
                                                                                    also see:  xtreg postestimation
-----------------------------------------------------------------------------------------------------------------------

Title

[XT] xtreg -- Fixed-, between-, and random-effects, and population-averaged linear models

Syntax

GLS random-effects (RE) model

      xtreg depvar [indepvars] if [, re RE_options]

Between-effects (BE) model

      xtreg depvar [indepvars] if , be [BE_options]

Fixed-effects (FE) model

      xtreg depvar [indepvars] if , fe [FE_options]

ML random-effects (MLE) model

      xtreg depvar [indepvars] if [weight] , mle [MLE_options]

Population-averaged (PA) model

      xtreg depvar [indepvars] if [weight] , pa [PA_options]

RE_options          description
-----------------------------------------------------------------------------------------------------------------
Model
   i(varname_i)       use varname_i as the panel ID variable
   re                use random-effects estimator; the default
   sa                use Swamy-Arora estimator of the variance components

SE/Robust
   vce(vcetype)       vcetype may be robust, bootstrap or jackknife
   robust             synonym for vce(robust)
   cluster(varname) adjust standard errors for intragroup correlation
   nonest             do not check that panels are nested within clusters

Reporting
   level(#)          set confidence level; default is level(95)
   theta             report theta

-----------------------------------------------------------------------------------------------------------------

BE_options          description
-----------------------------------------------------------------------------------------------------------------
Model
   i(varname_i)       use varname_i as the panel ID variable
   be                use between-effects estimator
   wls                use weighted least squares

SE
   vce(vcetype)       vcetype may be bootstrap or jackknife

Reporting
   level(#)          set confidence level; default is level(95)
-----------------------------------------------------------------------------------------------------------------

FE_options          description
-----------------------------------------------------------------------------------------------------------------
Model
   i(varname_i)       use varname_i as the panel ID variable
   fe                use fixed-effects estimator

SE/Robust
   vce(vcetype)       vcetype may be robust, bootstrap or jackknife
   robust             synonym for vce(robust)
   cluster(varname) adjust standard errors for intragroup correlation
   nonest             do not check that panels are nested within clusters
   dfadj             adjust the cluster-robust VCE for the within transform; seldom used

Reporting
   level(#)          set confidence level; default is level(95)
-----------------------------------------------------------------------------------------------------------------

Description

xtreg fits cross-sectional time-series regression models.  In particular, xtreg with the be option fits
random-effects models using the between regression estimator; with the fe option, it fits fixed-effects models
(using the within regression estimator); and with the re option, it fits random-effects models using the GLS
estimator (producing a matrix-weighted average of the between and within results).  See xtdata for a faster way
to fit fixed- and random-effects models.

Options for BE model

      +-------+
----+ Model +----------------------------------------------------------------------------------------------------

i(varname_i); see estimation options.

be requests the between regression estimator.

wls specifies that, in the case of unbalanced data, weighted least squares be used rather than the default OLS.
      Both methods produce consistent estimates.

      +----+
----+ SE +-------------------------------------------------------------------------------------------------------

vce(vcetype); see vce_option

      +-----------+
----+ Reporting +------------------------------------------------------------------------------------------------

level(#); see estimation options.

Options for FE model

      +-------+
----+ Model +----------------------------------------------------------------------------------------------------

i(varname_i); see estimation options.

fe requests the fixed-effects (within) regression estimator.

      +-----------+
----+ SE/Robust +------------------------------------------------------------------------------------------------

vce(vcetype); see vce_option.

robust, cluster(varname); see estimation options.

nonest removes the check that the panels are nested within clusters.  The cluster-robust VCE generally assumes
      that the panels are nested within the clusters, or that there are many observations per panel.

dfadj adjusts the cluster-robust VCE for the within transform.  dfadj will produce a conservative VCE when panels are not nested within clusters, even when there are only a few observations per panel.

      +-----------+
----+ Reporting +------------------------------------------------------------------------------------------------

level(#); see estimation options.

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

扫码加我 拉你入群

扫码加我 拉你入群

扫码加我 拉你入群

扫码加我 拉你入群

扫码加我 拉你入群

扫码加我 拉你入群

扫码加我 拉你入群

分享

扫码加好友，拉您进群

扫码加我拉你入群

扫码加我拉你入群

扫码加我拉你入群

扫码加我拉你入群

扫码加我拉你入群

扫码加我拉你入群

扫码加我拉你入群