------------------------------------------------------------------------------
y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
x1 | -.7483883 .2033164 -3.68 0.000 -1.146924 -.3498526
x2 | .109645 .0093469 11.73 0.000 .0913235 .1279665
_cons | 1.468996 .1144937 12.83 0.000 1.244568 1.693423
-------------+----------------------------------------------------------------
sigma_u | .45126578
sigma_e | 12.081701
rho | .00139317 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(92, 11251) = 0.17 Prob > F = 1.0000
得到的结果是要看最后一行的F检验吗?
这个——F test that all u_i=0: F(92, 11251) = 0.17 Prob > F = 1.0000
如果是看这个的话是不能使用固定效应模型的吧?
而且这个P值为1好诡异啊。。。
但是我作hausman检验,结果又是是这样
. hausman fe
---- Coefficients ----
| (b) (B) (b-B) sqrt(diag(V_b-V_B))
| fe re Difference S.E.
-------------+----------------------------------------------------------------
x1 | -.7483883 -.74597 -.0024184 .0170311
x2 | .109645 .1111525 -.0015075 .0008628
------------------------------------------------------------------------------
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
你这个豪斯曼检验有误啊,检验命令是hausman FE RE,constant sigmamore
所以你得先分别做fe和re的回归,再进行豪斯曼检验。另外,fe回归默认就是把固定效应和混合效应作比较,结果不显著表明固定效应更好,而re回归默认就是把随机效应和混合效应作比较,结果不显著表明随机效应更好