Annexe (5)
22
Tableau (3) : Modèle MCO
23
Annexe (6)
Tableau (4) : Modèle d'effet fixe
(muette)
|
SS
|
df MS
|
|
Number of obs
|
|
|
|
|
|
3517.13667
|
51 68.9634641
|
|
Prob > F
|
|
|
1185.58503
|
629 1.88487286
|
|
R-squared
|
|
|
|
|
|
Adj R-squared
|
|
|
|
|
|
4702.7217
|
680 6.9157672
|
|
Root MSE
|
|
|
Source
|
|
|
P>|t|
|
|
|
|
-.0000637
|
8.18e-06 -7.79
|
0.000
|
F( 51, 629)
-.0000797
|
|
|
Model
|
-.0080074
|
.0028437 -2.82
|
0.005
|
-.0135916
|
|
|
Residual
|
-.0047589
|
.0080988 -0.59
|
0.557
|
-.0206629
|
|
|
.1512757
|
.0146852 10.30
|
0.000
|
.1224378
|
|
|
Total
|
-.4781322
|
.0914723 -5.23
|
0.000
|
-.6577603
|
|
|
-.4957068
|
.0795044 -6.23
|
0.000
|
-.651833
|
|
|
-.177605
|
.0752474 -2.36
|
0.019
|
-.3253716
|
|
|
tpib
|
Coef.
-.0143376
|
Std. Err. t
.0048823 -2.94
|
0.003
|
[95% Conf.
-.0239252
|
|
24
Annexe (7)
Tableau (5) : Modèle d'effets fixes
(Within)
|
Fixed-effects (within) regression
Group variable: id
|
|
|
|
|
|
|
t
|
P>|t|
|
|
|
|
R-sq: within = 0.3591
|
between = 0.3733 overall =
0.3434
-.0000637
-.0080074
-.0047589
.1512757
-.4781322
-.4957068
-.177605
-.0143376
8.714733
|
8.18e-06 .0028437 .0080988
.0146852 .0914723 .0795044 .0752474
.0048823 .8402509
|
-7.79 -2.82 -0.59
10.30 -5.23 -6.23 -2.36
-2.94 10.37
|
0.000 0.005 0.557
0.000 0.000 0.000 0.019
0.003 0.000
|
Number of obs =
Number of groups =
Obs per group: min =
avg = max =
F(8,629) =
-.0000797
-.0135916
-.0206629
.1224378
-.6577603
-.651833
-.3253716
-.0239252
7.064696
|
|
Annexe (8)
Tableau (6) : Modèle d'effet
aléatoire
25
26
Annexe (9)
Tableau (7) : Test d'Hausman
. hausman fe re
Note: the rank of the differenced variance matrix (7) does
not equal the number of coefficients being tested (8); 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.
|
|
(b-B)
Difference
|
|
|
(b) (B)
fe re
|
|
-.0000637
|
-.0000597
|
-4.02e-06
|
|
|
-.0080074
|
-.0072992
|
-.0007081
|
|
|
-.0047589
|
-.0011403
|
-.0036185
|
|
|
.1512757
|
.1444938
|
.0067819
|
|
|
-.4781322
|
-.3928342
|
-.085298
|
|
|
-.4957068
|
-.4318874
|
-.0638194
|
|
|
-.177605
|
-.0413254
|
-.1362796
|
|
|
-.0143376
|
-.0107395
|
-.0035981
|
|
sqrt(diag(V_b-V_B))
S.E.
Coefficients
3.12e-06
.0013557
.0002255
.0039583
.0456137
.0387827
.0534109
.0031563 b = consistent under Ho and
Ha; obtained from xtreg B = inconsistent under Ha,
efficient under Ho; obtained from xtreg cpib
det
infl
inv
depmil
depsa
dem
touvnet
Test: Ho: difference in coefficients not systematic
chi2(7) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 23.91
Prob>chi2 = 0.0012
(V_b-V_B is not positive definite)
.
Breusch and Pagan Lagrangian multiplier test for random
effects
tpib[id,t] = Xb + u[id] +
e[id,t]
|
|
|
|
|
Estimated results
|
Var
1.884873
chibar2(01)
|
sd
|
|
= sqrt(Var)
6.915767
3.130354
=
2.629785 1.372907 1.769281
1232.61
Prob > chibar2 = 0.0000
| 
