Annexe 3 : Résultats des tests
Les estimations et tests sont faits sur le logiciel Small STATA
8.2. 1- Statistiques descriptives
Variable
|
| Mean
+
|
Std. Dev.
|
Min
|
Max | Observations
+
|
m_banc
|
overall |
|
94.81657
|
2.747193
|
88.85
|
99.65 |
|
N
|
=
|
35
|
|
between |
|
|
2.743491
|
90.268
|
99.174 |
|
n
|
=
|
7
|
|
within |
|
|
.951749
|
92.29057
|
96.41057 |
|
T
|
=
|
5
|
pib_hbt
|
overall |
|
262.4714
|
105.0595
|
113.7
|
446.2 |
|
N
|
=
|
35
|
|
between |
|
|
110.3224
|
121.06
|
435.78 |
|
n
|
=
|
7
|
|
within |
|
|
17.273
|
225.2714
|
302.0714 |
|
T
|
=
|
5
|
t_pgr_b
|
overall |
|
3.147714
|
8.647716
|
-12.5
|
28.57 |
|
N
|
=
|
35
|
|
between |
|
|
4.70726
|
-2.952
|
11.294 |
|
n
|
=
|
7
|
|
within |
|
|
7.431794
|
-20.64629
|
20.42371 |
|
T
|
=
|
5
|
b_petit
|
overall |
|
59.71257
|
17.79981
|
33.33
|
100 |
|
N
|
=
|
35
|
|
between |
|
|
17.3519
|
43.702
|
91.556 |
|
n
|
=
|
7
|
|
within |
|
|
7.153115
|
45.93657
|
70.98057 |
|
T
|
=
|
5
|
vol
_crd
|
overall |
|
.5411716
|
.4597332
|
.071824
|
1.64283 |
|
N
|
=
|
35
|
|
between |
|
|
.4773749
|
.0954074
|
1.469874 |
|
n
|
=
|
7
|
|
within |
|
|
.1013806
|
.3537843
|
.8251508 |
|
T
|
=
|
5
|
v_crimf
|
overall |
|
.0272047
|
.0203366
|
.001866
|
.075865 |
|
N
|
=
|
35
|
|
between |
|
|
.018235
|
.0026008
|
.0575586 |
|
n
|
=
|
7
|
|
within |
|
|
.0109627
|
-.0044269
|
.0586213 |
|
T
|
=
|
5
|
2 - Matrice de corrélation
|
|
|
|
|
|
|
| m_banc pib_hbt
|
t_pgr_b
|
b_petit
|
vol_crd v_crimf
|
|
|
+
m_banc |
pib_hbt |
|
t_pgr_b |
|
b_petit |
|
vol _crd |
|
v_crimf |
|
|
1.0000 -0.7847* 0.0000 0.2575 0.1353 0.6165* 0.0001 -0.7931*
0.0000 -0.2245 0.1947
|
1.0000
-0.0767 0.6616 -0.7198* 0.0000 0.8981* 0.0000 0.4804*
0.0035
|
1.0000
0.0637 0.7162 -0.1690 0.3318 0.5053* 0.0020
|
1.0000
-0.6486* 0.0000 -0.5410* 0.0008
|
1.0000
0.2209 0.2023
|
1.0000
|
|
3 - Test de présence d'effets individuels
Fixed-effects (within) regression Number of obs = 35
Group variable (i) : crossid Number of groups = 7
R-sq: within = 0.6178 Obs per group: min = 5
between = 0.3327 avg = 5.0
overall = 0.3448 max = 5
F(5,23) = 7.43
corr(u_i, Xb) = -0.6518 Prob > F = 0.0003
m_banc | Coef. Std. Err. t P>|t| [95% Conf. Interval]
pib_hbt
|
|
|
-.0433641
|
.0107511
|
|
-4.03 0.001
|
|
-.0656045
|
-.0211237
|
b_petit
|
|
|
.0190392
|
.0195189
|
|
0.98 0.339
|
|
-.0213388
|
.0594171
|
t_pgr_b
|
|
|
.0162727
|
.0206284
|
|
0.79 0.438
|
|
-.0264004
|
.0589458
|
vol _crd
|
|
|
5.501833
|
2.348176
|
|
2.34 0.028
|
|
.6442602
|
10.35941
|
v_crimf
|
|
|
-47.16463
|
21.24999
|
|
-2.22 0.037
|
|
-91.12359
|
-3.205668
|
_cons
|
|
|
103.316
|
2.832919
|
|
36.47 0.000
|
|
97.45564
|
109.1763
|
sigma_u
|
|
|
3.0104067
|
|
|
|
|
|
|
sigma_e
|
|
|
.71541673
|
|
|
|
|
|
|
rho
|
|
|
.94654258
|
(fraction
|
of
|
variance due
|
to
|
u_i)
|
|
F test that
|
all
|
u_i=0:
|
F(6, 23) =
|
|
22.42
|
|
Prob >
|
F = 0.0000
|
|
4 - Test de spécification de Hausman (effet
fixe ou effet aléatoire)
---- Coefficients ----
| (b) (B) (b-B) sqrt (diag (V_b-V_B))
| sql . Difference S.E.
+
pib_hbt | -.0433641 -.0231721 -.020192 .0064676
t_pgr_b | .0162727 .0111748 .0050979 .
b_petit | .0190392 .0030578 .0159814 .
vol _crd | 5.501833 .2821059 5.219727 1.481768
v_crimf | -47.16463 -17.63811 -29.52652 11.35284
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)
= 19.13
Prob>chi2 = 0.0018
(V_b-V_B is not positive definite)
5 - Régression multiple (avec effets
fixes)
Fixed-effects (within) regression Number of obs = 35
Group variable (i) : crossid Number of groups = 7
R-sq: within = 0.6178 Obs per group: min = 5
between = 0.3327 avg = 5.0
overall = 0.3448 max = 5
F(5,23) = 7.43
corr(u_i, Xb) = -0.6518 Prob > F = 0.0003
m_banc | Coef. Std. Err. t P>|t| [95% Conf. Interval]
pib_hbt
|
|
|
-.0433641
|
.0107511
|
|
-4.03 0.001
|
|
-.0656045
|
-.0211237
|
t_pgr_b
|
|
|
.0162727
|
.0206284
|
|
0.79 0.438
|
|
-.0264004
|
.0589458
|
b_petit
|
|
|
.0190392
|
.0195189
|
|
0.98 0.339
|
|
-.0213388
|
.0594171
|
vol _crd
|
|
|
5.501833
|
2.348176
|
|
2.34 0.028
|
|
.6442602
|
10.35941
|
v_crimf
|
|
|
-47.16463
|
21.24999
|
|
-2.22 0.037
|
|
-91.12359
|
-3.205668
|
_cons
|
|
|
103.316
|
2.832919
|
|
36.47 0.000
|
|
97.45564
|
109.1763
|
sigma_u
|
|
|
3.0104067
|
|
|
|
|
|
|
sigma_e
|
|
|
.71541673
|
|
|
|
|
|
|
rho
|
|
|
.94654258
|
(fraction
|
of
|
variance due
|
to
|
u_i)
|
|
F test that
|
all
|
u_i=0:
|
F(6, 23) =
|
|
22.42
|
|
Prob >
|
F = 0.0000
|
|
6 - Test de normalisation des erreurs
Skewness/Kurtosis tests for Normality
joint
Variable | Pr(Skewness) Pr(Kurtosis) adj chi2(2)
Prob>chi2 +
residu | 0.788 0.023 5.12 0.0773
7 - Test d'autocorrélation des erreurs
Regression with robust standard errors Number of obs = 28
F( 5, 6) = 44.24
Prob > F = 0.0001
R-squared = 0.3795
Number of clusters (crossid) = 7 Root MSE = 1.0965
|
|
|
Robust
|
|
|
|
|
D.m_banc
pib_hbt
|
|
|
|
Coef.
|
Std. Err.
|
t
|
P>|t|
|
[95% Conf.
|
Interval]
|
t_pgr_b
|
D1 |
|
|
-.0501591
|
.0164165
|
-3.06
|
0.022
|
-.0903287
|
-.0099894
|
b_petit
|
D1 |
|
|
.0122495
|
.0152547
|
0.80
|
0.453
|
-.0250773
|
.0495763
|
vol _crd
|
D1 |
|
|
.0369609
|
.009612
|
3.85
|
0.009
|
.0134413
|
.0604805
|
v_crimf
|
D1 |
|
|
5.786839
|
2.215428
|
2.61
|
0.040
|
.3658811
|
11.2078
|
|
D1 |
|
-42.80981
|
18.12651
|
-2.36
|
0.056
|
-87.16378
|
1.544166
|
|
Wooldridge test for autocorrelation in panel data H0: no
first-order autocorrelation
F( 1, 6) = 0.265
Prob > F = 0.6250
Les déterminants de la faible
8 - Test d'homoscédasticité
|
bancarisation dans l'UEMOA
explicatives
|
|
Régression du
|
carré du résidu sur les variables
|
|
SS
|
df MS
|
Number of obs
|
= 35
|
+
|
|
|
F( 5, 29)
|
=29526.49
|
Model |
|
14658438.2
|
5 2931687.65
|
Prob > F
|
= 0.0000
|
Residual |
|
2879.41238
|
29 99.2900822
|
R-squared
|
= 0.9998
|
+
|
|
|
Adj R-squared
|
= 0.9998
|
Total |
|
14661317.7
|
34 431215.225
|
Root MSE
|
= 9.9644
|
red |
|
Coef.
|
Std. Err. t
|
P>|t| [95% Conf.
|
Interval]
|
pib_hbt |
|
-8.263501
|
.0517883 -159.56
|
0.000 -8.36942
|
-8.157582
|
b_petit |
|
4.071197
|
.1592721 25.56
|
0.000 3.745449
|
4.396945
|
t_pgr_b |
|
2.995623
|
.2720108 11.01
|
0.000 2.439298
|
3.551948
|
vol _crd |
|
1054.956
|
10.7014 98.58
|
0.000 1033.069
|
1076.842
|
v_crimf |
|
-8601.311
|
165.7161 -51.90
|
0.000 -8940.239
|
-8262.384
|
_cons |
|
10581.27
|
14.58328 725.58
|
0.000 10551.45
|
10611.1
|
|
Test d'hétéroscédasticité
inter-individuelle
Modified Wald test for groupwise heteroskedasticity
in fixed effect regression model
H0: sigma(i)^2 = sigma^2 for all i
chi2 (7) = 158.06
Prob>chi2 = 0.0000
|