hausman检验结果如下，改用固定效应模型还是。。。急

9343

收藏 2014-03-12

悬赏 5 个论坛币未解决

hausman检验结果如下，该用固定效应模型还是随机效应模型？采用后应该做哪些检验，请写上操作命令语句以及检验结果判定方法，灰常感谢！
b = consistent under Ho and Ha; obtained from xtreg
         B = inconsistent under Ha, efficient under Ho; obtained from xtreg

Test:  Ho:  difference in coefficients not systematic

               chi2(0) = (b-B)'[(V_b-V_B)^(-1)](b-B)
                        =       0.00
            Prob>chi2 =          .
            (V_b-V_B is not positive definite)

本人做完分析后看到t，P>t和P>|z|不知什么意思是，该系数是显著还是不显著？
xtreg  y x1 x2 x3,fe    *固定模型

Fixed-effects (within) regression             Number of obs    =       45
Group variable: code                         Number of groups =       9

R-sq:  within  = 0.0593                      Obs per group: min =       5
   between = 0.9984                                     avg =    5.0
   overall = 0.9808                                     max =       5

                                             F(3,33)          =    0.69
corr(u_i, Xb)  = 0.9945                      Prob > F          = 0.5626

------------------------------------------------------------------------------
         y |    Coef. Std. Err.    t P>|t|    [95% Conf. Interval]
-------------+----------------------------------------------------------------
      x1 | .0011169 .0067139    0.17 0.869 -.0125425 .0147764
      x2 | 86.30024 60.53457    1.43 0.163 -36.85827 209.4588
      x3 | .0053639 .1217428    0.04 0.965 -.2423236 .2530514
   _cons | 15.94421 7.107634    2.24 0.032    1.483615    30.4048
-------------+----------------------------------------------------------------
   sigma_u |  6.6783079
   sigma_e |  1.4320673
      rho | .9560388 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0:    F(8, 33) =    1.13             Prob > F = 0.3725

xtreg y x1 x2 x3,re          *随机模型

Random-effects GLS regression                Number of obs    =       45
Group variable: code                         Number of groups =       9

R-sq:  within  = 0.0585                      Obs per group: min =       5
   between = 0.9987                                     avg =    5.0
   overall = 0.9810                                     max =       5

Random effects u_i ~ Gaussian                Wald chi2(3)    = 2120.52
corr(u_i, X)    = 0 (assumed)             Prob > chi2       = 0.0000

------------------------------------------------------------------------------
         y |    Coef. Std. Err.    z P>|z|    [95% Conf. Interval]
-------------+----------------------------------------------------------------
      x1 | .0010005    .00668    0.15 0.881 -.0120921    .014093
      x2 | 234.8571 6.09749 38.52 0.000    222.9062 246.8079
      x3 |  -.0165454 .0540021 -0.31 0.759 -.1223876 .0892968
   _cons | -.274155 2.590904 -0.11 0.916 -5.352233 4.803923
-------------+----------------------------------------------------------------
   sigma_u |       0
   sigma_e |  1.4320673
      rho |       0 (fraction of variance due to u_i)
-----------------------------------------------------------------------------

顺便请问模型设定是否合理？

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

ywh19860616

2014-3-12 18:40:05

hausman检验矩阵出现了非正定，你加上
sigmamore 或者 sigmaless  选项试试。

sigmamore             base both (co)variance matrices on disturbance variance estimate from efficient estimator
sigmaless             base both (co)variance matrices on disturbance variance estimate from consistent estimator

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

brave_try

2014-3-12 18:50:49

ywh19860616 发表于 2014-3-12 18:40
hausman检验矩阵出现了非正定，你加上
sigmamore 或者 sigmaless 选项试试。

hausman fe,sigmamore

Note: the rank of the differenced variance matrix (0) does not equal the number of coefficients being tested (3); be sure this is
      what you expect, or there may be problems computing the test.  Examine the output of your estimators for anything unexpected
      and possibly consider scaling your variables so that the coefficients are on a similar scale.

               ---- Coefficients ----
         |    (b)       (B)          (b-B)    sqrt(diag(V_b-V_B))
         |    fe          re       Difference       S.E.
-------------+----------------------------------------------------------------
      x1 | .0010005    .0010005             0             0
      x2 | 234.8571    234.8571             0             0
      x3 | -.0165454 -.0165454             0             0
------------------------------------------------------------------------------
                        b = consistent under Ho and Ha; obtained from xtreg
         B = inconsistent under Ha, efficient under Ho; obtained from xtreg

Test:  Ho:  difference in coefficients not systematic

               chi2(0) = (b-B)'[(V_b-V_B)^(-1)](b-B)
                        =       0.00
            Prob>chi2 =          .
            (V_b-V_B is not positive definite)

hausman fe,sigmaless

Note: the rank of the differenced variance matrix (0) does not equal the number of coefficients being tested (3); be sure this is
      what you expect, or there may be problems computing the test.  Examine the output of your estimators for anything unexpected
      and possibly consider scaling your variables so that the coefficients are on a similar scale.

               ---- Coefficients ----
         |    (b)       (B)          (b-B)    sqrt(diag(V_b-V_B))
         |    fe          re       Difference       S.E.
-------------+----------------------------------------------------------------
      x1 | .0010005    .0010005             0             0
      x2 | 234.8571    234.8571             0             0
      x3 | -.0165454 -.0165454             0             0
------------------------------------------------------------------------------
                        b = consistent under Ho and Ha; obtained from xtreg
         B = inconsistent under Ha, efficient under Ho; obtained from xtreg

Test:  Ho:  difference in coefficients not systematic

               chi2(0) = (b-B)'[(V_b-V_B)^(-1)](b-B)
                        =       0.00
            Prob>chi2 =          .
            (V_b-V_B is not positive definite)

.

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

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