Annexe 7 : résultats des tests d'hausman pour
les modèles
. hausman fixed
---- Coefficients ----
(b) (B)
fixed .
|
(b-B) Difference
|
sqrt(diag(V_b-V_B)) S.E.
|
ASP
|
| -.1040167
|
-.0856388
|
-.018378
|
.0042558
|
lnTG
|
| -.0294086
|
-.0247023
|
-.0047063
|
.0013607
|
lnOC
|
| .0256118
|
.0203253
|
.0052865
|
.0038075
|
FDI
|
| .0006274
|
.0006088
|
.0000186
|
.
|
PFI
|
| .0384899
|
.041615
|
-.0031251
|
.
|
STI
|
| -.0282517
|
-.0269044
|
-.0013473
|
.
|
CPIB
|
| .0018299
|
.0021477
|
-.0003178
|
.
|
VTCER
|
| -.0114757
|
-.0123185
|
.0008428
|
.
|
VPIB
|
| 6.35e-06
|
-.0000292
|
.0000356
|
7.06e-06
|
VCTDE
|
| -.0029047
|
-.0104178
|
.0075131
|
.
|
VINF
|
| -3.68e-07
|
-2.05e-07
|
-1.62e-07
|
.
|
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(10)
|
=
|
(b-B)'[(V_b-V_B)^(-1)](b-B)
|
|
=
|
26.21
|
Prob>chi2
|
=
|
0.0035
|
. hausman fixes
---- Coefficients ----
(b) (B)
fixes .
|
(b-B) Difference
|
sqrt(diag(V_b-V_B)) S.E.
|
MCPIB
|
| .0068244
|
.0005831
|
.0062413
|
.0108725
|
MlnTG
|
| .2034575
|
.1336621
|
.0697954
|
.0778277
|
MlnOC
|
| .0354035
|
.00482
|
.0305835
|
.1176474
|
FERT
|
| -1.734617
|
-2.275468
|
.5408514
|
.5440932
|
MSTI
|
| -.6018115
|
-.6139975
|
.012186
|
.2097658
|
VINF
|
| .0000657
|
.0000529
|
.0000128
|
.0000127
|
VCTDE
|
| .4322402
|
.3189378
|
.1133024
|
.2972571
|
VPIB
|
| .0034928
|
.0034407
|
.0000522
|
.0011709
|
VTCER
|
| -.1543973
|
-.231171
|
.0767737
|
.1751779
|
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(8) =
(b-B)'[(V_b-V_B)^(-1)](b-B)
= 3.39
Prob>chi2 = 0.9079
Annexe 8 : Résultats des tests
d'hétéroscédasticité sur les modèles
.Modified Wald test for groupwise heteroskedasticity in
fixed effect regression model
H0: sigma(i)^2 = sigma^2 for all
i
chi2 (18) = 693.34
Prob>chi2 = 0.0000
Cross-sectional time-series FGLS regression
Coefficients: generalized least squares Panels:
heteroskedastic
Correlation: no autocorrelation
Estimated
|
covariances
|
=
|
17
|
Number of obs =
|
100
|
Estimated
|
autocorrelations
|
=
|
0
|
Number of groups =
|
17
|
Estimated
|
coefficients
|
=
|
7
|
Obs per group: min =
|
5
|
|
|
|
|
avg =
|
5.882353
|
|
|
|
|
max =
|
6
|
|
|
|
|
Wald chi2(6) =
|
1515.57
|
Log likelihood
|
=
|
-138.7242
|
Prob > chi2 =
|
0.0000
|
H | Coef.
|
Std. Err.
|
z
|
P>|z|
|
[95% Conf.
|
Interval]
|
MCPIB
|
| -.0058625
|
.015304
|
-0.38
|
0.702
|
-.0358578
|
.0241329
|
MlnTG
|
| .9455804
|
.1672175
|
5.65
|
0.000
|
.6178402
|
1.273321
|
MlnOC
|
| 1.111018
|
.1235995
|
8.99
|
0.000
|
.8687677
|
1.353269
|
MDFI
|
| .6551396
|
.3391258
|
1.93
|
0.053
|
-.0095348
|
1.319814
|
POPRAL
|
| -.0456649
|
.0028193
|
-16.20
|
0.000
|
-.0511906
|
-.0401392
|
VPIB
|
| -.0052023
|
.0023434
|
-2.22
|
0.026
|
-.0097953
|
-.0006094
|
_cons | 9.766002
|
.3960112
|
24.66
|
0.000
|
8.989834
|
10.54217
|
.
. estimates store hetero
.
. xtgls H MCPIB MlnTG MlnOC MDFI POPRAL VPIB
Cross-sectional time-series FGLS regression
Coefficients: generalized least squares Panels:
homoskedastic
Correlation: no autocorrelation
Estimated
|
covariances
|
=
|
1
|
Number of obs =
|
100
|
Estimated
|
autocorrelations
|
=
|
0
|
Number of groups =
|
17
|
Estimated
|
coefficients
|
=
|
7
|
Obs per group: min =
|
5
|
|
|
|
|
avg =
|
5.882353
|
|
|
|
|
max =
|
6
|
|
|
|
|
Wald chi2(6) =
|
125.24
|
Log likelihood
|
=
|
-174.1838
|
Prob > chi2 =
|
0.0000
|
H | Coef.
|
Std. Err.
|
z
|
P>|z|
|
[95% Conf.
|
Interval]
|
MCPIB
|
| .1210946
|
.0562855
|
2.15
|
0.031
|
.0107771
|
.2314121
|
MlnTG
|
| 1.436827
|
.4658322
|
3.08
|
0.002
|
.5238123
|
2.349841
|
MlnOC
|
| .7761341
|
.3586263
|
2.16
|
0.030
|
.0732395
|
1.479029
|
MDFI
|
| 2.22052
|
1.68022
|
1.32
|
0.186
|
-1.07265
|
5.51369
|
POPRAL
|
| -.0665186
|
.0095558
|
-6.96
|
0.000
|
-.0852476
|
-.0477896
|
VPIB
|
| -.015764
|
.00809
|
-1.95
|
0.051
|
-.0316201
|
.0000921
|
_cons | 12.35888
|
1.176686
|
10.50
|
0.000
|
10.05261
|
14.66514
|
.
. local df = e(N_g) - 1
.
. lrtest hetero . , df(`df')
Likelihood-ratio test LR chi2(16) = 70.92
(Assumption: . nested in hetero) Prob > chi2 =
0.0000
|