我的固定效应和hausman检验结果做出来是以下结果,想向高手请教一下这是有什么问题么? 是变量本身就不好还是可以不做hausman检验直接做随即效应可以了?
谢谢。
. xtreg innovation train machine concel contracting domestic export,fe
Fixed-effects (within) regression Number of obs = 24
Group variable: region Number of groups = 3
R-sq: within = 0.9411 Obs per group: min = 8
between = 0.9997 avg = 8.0
overall = 0.9901 max = 8
F(6,15) = 39.97
corr(u_i, Xb) = 0.8067 Prob > F = 0.0000
------------------------------------------------------------------------------
innovation | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
train | .2955226 .185245 1.60 0.131 -.0993178 .690363
machine | .047459 .0428804 1.11 0.286 -.0439385 .1388565
concel | -.1032061 .1770827 -0.58 0.569 -.480649 .2742367
contracting | .6621216 .1496214 4.43 0.000 .3432112 .981032
domestic | .0299726 .0144563 2.07 0.056 -.0008403 .0607855
export | .0567115 .0598268 0.95 0.358 -.0708063 .1842293
_cons | -3.344421 1.614725 -2.07 0.056 -6.786126 .0972847
-------------+----------------------------------------------------------------
sigma_u | .04096643
sigma_e | .11656366
rho | .10993854 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(2, 15) = 0.13 Prob > F = 0.8792
. hausman fe
---- Coefficients ----
| (b) (B) (b-B) sqrt(diag(V_b-V_B))
| fe re Difference S.E.
-------------+----------------------------------------------------------------
train | .2955226 .2501052 .0454174 .1194769
machine | .047459 .0437281 .0037309 .0167939
concel | -.1032061 -.0574301 -.0457761 .1145066
contracting | .6621216 .6910648 -.0289433 .099885
domestic | .0299726 .0297878 .0001848 .0047293
export | .0567115 .0547764 .0019351 .0225565
------------------------------------------------------------------------------
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(6) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 0.21
Prob>chi2 = 0.9998