stata计量高手请进

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

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

分享0 收藏1 回帖

关键词：Stata tata absorb 是什么原因 xtreg absorb

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

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

mzdg 发表于 2017-3-23 17:31
非常感谢

You are welcome.

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。

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！

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.