请教：面板数据计量结果是否可用

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

 下述结果是用面板数据做出的结果，是否可用？如不可用需如何改进？

谢谢

xtreg v12 log v7 v10  sizeage, fe

Fixed-effects (within) regression               Number of obs      =       250
Group variable (i): stkcd                       Number of groups   =        50

R-sq:  within  = 0.1516                         Obs per group: min =         5
       between = 0.2099                                        avg =       5.0
       overall = 0.1268                                        max =         5

                                                F(3,197)           =     11.74
corr(u_i, Xb)  = -0.7248                        Prob > F           =    0.0000

------------------------------------------------------------------------------
         v12 |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         log |   658.4801   289.9299     2.27   0.024     86.71535    1230.245
          v7 |   141.7374   33.28625     4.26   0.000     76.09428    207.3805
         v10 |  (dropped)
     sizeage |  -82.85968   32.44082    -2.55   0.011    -146.8355   -18.88383
       _cons |  -1200.356   295.1242    -4.07   0.000    -1782.365    -618.348
-------------+----------------------------------------------------------------
     sigma_u |  19.655635
     sigma_e |  20.578124
         rho |  .47708382   (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0:     F(49, 197) =     1.74             Prob > F = 0.0043

 xtgls v12 log v7 v10,panel(he)

Cross-sectional time-series FGLS regression

Coefficients:  generalized least squares
Panels:        heteroskedastic
Correlation:   no autocorrelation

Estimated covariances      =        50          Number of obs      =       250
Estimated autocorrelations =         0          Number of groups   =        50
Estimated coefficients     =         4          Time periods       =         5
                                                Wald chi2(3)       =     98.41
Log likelihood             = -1019.364          Prob > chi2        =    0.0000

------------------------------------------------------------------------------
         v12 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         log |  -41.35655   4.553634    -9.08   0.000    -50.28151   -32.43159
          v7 |   13.37212   2.121515     6.30   0.000     9.214027    17.53021
         v10 |   2.608105   1.382795     1.89   0.059    -.1021235    5.318334
       _cons |  -78.79583   19.00019    -4.15   0.000    -116.0355   -41.55614
------------------------------------------------------------------------------

[此贴子已经被作者于2008-2-19 21:29:16编辑过]

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：面板数据 coefficients correlations fixed-effect correlation 数据 请教 面板 结果 计量

从变量的显著性来讲模型还是可以的，也就是说通过了统计检验，但是经济检验还需要进行，即系数的合理性