Annexe 4 : Test de spécification de Hausman
. xtreg chomage inflation pibreel investtotal txinteretreel
txdechangereel popact, fe
Fixed-effects (within) regression Number of obs
= 174 Group variable: id Number of groups
= 6
R-sq: Obs per group:
within = 0.1229
min = 29 between = 0.1730
avg = 29.0 overall = 0.1488
max = 29
F(6,162)
= 3.78 corr(u_i, Xb) = -0.5424 Prob > F
= 0.0015
|
chomage
|
Coef. Std. Err. t P>|t| [95% Conf.
Interval]
|
|
inflation
|
-.0138686 .0131712 -1.05 0.294 -.0398779
.0121408
|
|
pibreel
|
-.002417 .0086742 -0.28 0.781 -.0195461
.0147121
|
|
investtotal
|
-.0015792 .0064148 -0.25 0.806 -.0142467
.0110882
|
|
txinteretreel
|
-.1511477 .0609788 -2.48 0.014 -.2715635
-.0307318
|
|
txdechangereel
|
-.0240256 .0104966 -2.29 0.023 -.0447535
-.0032977
|
|
popact
|
.1606087 .0386909 4.15 0.000 .084205
.2370123
|
|
_cons
|
.254408 2.363055 0.11 0.914 -4.411955
4.920771
|
|
sigma_u
|
8.6508356
|
|
sigma_e
|
1.024694
|
|
rho
|
.98616366 (fraction of variance due to u_i)
|
F test that all u_i=0: F(5, 162) = 1078.46
Prob > F = 0.0000
. xtreg chomage inflation pibreel investtotal txinteretreel
txdechangereel popact, re
Random-effects GLS regression Number of obs
= 174
Group variable: id Number of
groups = 6
R-sq: Obs per
group:
within = 0.0007
min = 29 between = 0.6350
avg = 29.0 overall = 0.3659
max = 29
Wald chi2(6)
= 96.38 corr(u_i, X) = 0 (assumed) Prob > chi2
= 0.0000
|
chomage
|
Coef. Std. Err. z P>|z| [95% Conf.
Interval]
|
|
inflation
|
-.146962 .0743187 -1.98 0.048 -.2926239
-.0013
|
|
pibreel
|
-.2367551 .046209 -5.12 0.000 -.327323
-.1461872
|
|
investtotal
|
.2017864 .0293602 6.87 0.000 .1442416
.2593313
|
|
txinteretreel
|
.4805302 .2819672 1.70 0.088 -.0721153
1.033176
|
|
txdechangereel
|
-.0990087 .0593879 -1.67 0.095 -.2154069
.0173895
|
|
popact
|
-.2296692 .0519128 -4.42 0.000 -.3314164
-.1279221
|
|
_cons
|
18.88506 3.661218 5.16 0.000 11.7092
26.06091
|
|
sigma_u
|
0
|
|
sigma_e
|
1.024694
|
|
rho
|
0 (fraction of variance due to u_i)
|
. hausman fixed random, sigmamore
Note: the rank of the differenced variance matrix (5) does not
equal the number of coefficients being tested (6); 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)) fixed random Difference S.E.
|
|
inflation
|
-.0138686 -.146962 .1330934 .0157028
|
|
pibreel
|
-.002417 -.2367551 .2343381 .0191634
|
|
investtotal
|
-.0015792 .2017864 -.2033656 .022508
|
|
txinteretr~l
|
-.1511477 .4805302 -.6316779 .2101591
|
|
txdechange~l
|
-.0240256 -.0990087 .0749831 .0117295
|
|
popact
|
.1606087 -.2296692 .3902779 .2170115
|
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(5) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 162.13
Prob>chi2 = 0.0000
| 
