[面板数据求助] 混合模型和固定效应模型回归系数相反怎么解释 [推广有奖]

0关注
0粉丝

大专生

还不是VIP/贵宾

威望: 0 级
论坛币: 3615 个
通用积分: 0.0000
学术水平: 0 点
热心指数: 0 点
信用等级: 0 点
经验: 116 点
帖子: 3
精华: 0
在线时间: 91 小时
注册时间: 2017-2-20
最后登录: 2025-6-26

楼主

13864285961

发表于 2018-1-28 13:49:13 |AI写论文

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

本人用面板数据进行回归，解释变量中包含若干虚拟变量，所有虚拟变量均随时间有变化。理论预期因变量im与虚拟自变量wto之间有正相关关系。各界面的散点图和混合模型的结果均符合理论预期，但用固定效应回归时系数为负，应该则么解释这个现象，请大神赐教。.    reg im y v lp shk wto fta tbt

   Source |    SS    df    MS             Number of obs =    323
-------------+------------------------------          F(  7, 315) = 39.37
   Model |  7.33331247    7  1.04761607          Prob > F    =  0.0000
Residual |  8.38171622 315  .026608623          R-squared    =  0.4666
-------------+------------------------------          Adj R-squared =  0.4548
   Total |  15.7150287 322  .048804437          Root MSE    =  .16312

------------------------------------------------------------------------------
      im |    Coef. Std. Err.    t P>|t|    [95% Conf. Interval]
-------------+----------------------------------------------------------------
         y | .0333848 .0102369    3.26 0.001    .0132435 .0535262
         v | .0015776 .0025389    0.62 0.535 -.0034178    .006573
      lp | -.002643 .0004852 -5.45 0.000 -.0035977 -.0016883
      shk | .0017223 .012385    0.14 0.889 -.0226455 .0260901
      wto | .0773158 .0242243    3.19 0.002    .029654 .1249777
      fta | .1541093 .0328354    4.69 0.000    .0895049 .2187137
      tbt |  -.2451049 .0295673 -8.29 0.000 -.3032792 -.1869305
   _cons | .2311798 .054149    4.27 0.000    .1246405 .3377192
------------------------------------------------------------------------------

.    xtreg im y v lp tbt shk wto fta, fe /*Wald 检验*/

Fixed-effects (within) regression             Number of obs    =    323
Group variable: id                            Number of groups =       19

R-sq:  within  = 0.4982                      Obs per group: min =       17
   between = 0.2426                                     avg =    17.0
   overall = 0.3158                                     max =       17

                                             F(7,297)          =    42.12
corr(u_i, Xb)  = -0.3337                      Prob > F          = 0.0000

------------------------------------------------------------------------------
      im |    Coef. Std. Err.    t P>|t|    [95% Conf. Interval]
-------------+----------------------------------------------------------------
         y |  -.0285346 .0121858 -2.34 0.020 -.0525161 -.0045531
         v |  -.0126089 .0035742 -3.53 0.000 -.0196428 -.005575
      lp | .0014794 .000601    2.46 0.014    .0002966 .0026622
      tbt |  -.3463304 .0400708 -8.64 0.000    -.425189 -.2674717
      shk | .0224428 .0099703    2.25 0.025    .0028214 .0420642
      wto |  -.0492579 .0211784 -2.33 0.021 -.0909368 -.0075791
      fta | .1715934 .0254674    6.74 0.000    .1214741 .2217128
   _cons | .5768974 .0722429    7.99 0.000    .4347245 .7190702
-------------+----------------------------------------------------------------
   sigma_u |  .16835919
   sigma_e |  .10125961
      rho |  .73435352 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0:    F(18, 297) = 28.91          Prob > F = 0.0000

阿里旺旺图片20180128134907.jpg