ANNEXES
Annexe 1 :
Multicolinéarité
|
CS
|
EXP01
|
IDE
|
INF
|
INST
|
INST_CS
|
CREDIT
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
CS
|
1.000000
|
0.252161
|
0.099856
|
0.175612
|
-0.127173
|
0.966688
|
0.092555
|
|
EXP01
|
0.252161
|
1.000000
|
0.558814
|
-0.043593
|
-0.479200
|
0.130600
|
0.019055
|
|
IDE
|
0.099856
|
0.558814
|
1.000000
|
-0.032035
|
-0.108207
|
0.069199
|
0.071847
|
|
INF
|
0.175612
|
-0.043593
|
-0.032035
|
1.000000
|
-0.091216
|
0.152541
|
0.106804
|
|
INST
|
-0.127173
|
-0.479200
|
-0.108207
|
-0.091216
|
1.000000
|
0.122569
|
-0.031357
|
|
INST_CS
|
0.966688
|
0.130600
|
0.069199
|
0.152541
|
0.122569
|
1.000000
|
0.077661
|
|
CREDIT
|
0.092555
|
0.019055
|
0.071847
|
0.106804
|
-0.031357
|
0.077661
|
1.000000
|
Annexe 2 : Tests
préliminaires
· EXP
|
Null Hypothesis: Unit root (common unit root process)
|
|
|
Series: EXP01
|
|
|
|
|
|
|
Date: 05/07/21 Time: 10:47
|
|
|
|
|
|
Sample: 1980 2019
|
|
|
|
|
|
|
Exogenous variables: Individual effects
|
|
|
|
|
Automatic selection of maximum lags
|
|
|
|
|
|
Automatic lag length selection based on SIC: 0 to 1
|
|
|
|
Newey-West automatic bandwidth selection and Bartlett
kernel
|
|
|
Total number of observations: 193
|
|
|
|
|
|
Cross-sections included: 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Method
|
|
|
Statistic
|
|
Prob.**
|
|
|
Levin, Lin & Chu t*
|
|
-1.84652
|
|
0.0324
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
** Probabilities are computed assuming asympotic normality
|
|
|
|
|
|
|
|
|
|
|
Intermediate results on EXP01
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cross
|
2nd Stage
|
Variance
|
HAC of
|
|
Max
|
Band-
|
|
|
section
|
Coefficient
|
of Reg
|
Dep.
|
Lag
|
Lag
|
width
|
Obs
|
|
Gabon
|
-0.44702
|
43.628
|
14.465
|
0
|
9
|
13.0
|
39
|
|
Cameroun
|
-0.39280
|
12.867
|
3.9887
|
0
|
9
|
13.0
|
39
|
|
Congo
|
-0.19181
|
5.E+21
|
8.E+21
|
1
|
9
|
0.0
|
38
|
|
Guinée
|
-0.22863
|
36.705
|
12.164
|
1
|
9
|
5.0
|
38
|
|
Tchad
|
-0.10613
|
32.552
|
29.425
|
0
|
9
|
4.0
|
39
|
|
|
|
|
|
|
|
|
|
Coefficient
|
t-Stat
|
SE Reg
|
mu*
|
sig*
|
|
Obs
|
|
Pooled
|
-0.21718
|
-4.942
|
1.019
|
-0.539
|
0.857
|
|
193
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
· INF
|
Null Hypothesis: Unit root (common unit root process)
|
|
|
Series: INF
|
|
|
|
|
|
|
|
Date: 05/07/21 Time: 10:50
|
|
|
|
|
|
Sample: 1980 2019
|
|
|
|
|
|
|
Exogenous variables: Individual effects
|
|
|
|
|
Automatic selection of maximum lags
|
|
|
|
|
|
Automatic lag length selection based on SIC: 0 to 8
|
|
|
|
Newey-West automatic bandwidth selection and Bartlett
kernel
|
|
|
Total number of observations: 186
|
|
|
|
|
|
Cross-sections included: 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Method
|
|
|
Statistic
|
|
Prob.**
|
|
|
Levin, Lin & Chu t*
|
|
-8.02814
|
|
0.0000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
** Probabilities are computed assuming asympotic normality
|
|
|
|
|
|
|
|
|
|
|
Intermediate results on INF
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cross
|
2nd Stage
|
Variance
|
HAC of
|
|
Max
|
Band-
|
|
|
section
|
Coefficient
|
of Reg
|
Dep.
|
Lag
|
Lag
|
width
|
Obs
|
|
Gabon
|
-0.95916
|
48.212
|
8.0000
|
1
|
9
|
13.0
|
38
|
|
Cameroun
|
-0.74394
|
40.188
|
2.3717
|
0
|
9
|
31.0
|
39
|
|
Congo
|
-0.75118
|
45.759
|
3.0766
|
0
|
9
|
33.0
|
39
|
|
Guinée
|
-0.55797
|
9.2705
|
16.616
|
8
|
9
|
38.0
|
31
|
|
Tchad
|
-1.03012
|
86.670
|
4.4478
|
0
|
9
|
38.0
|
39
|
|
|
|
|
|
|
|
|
|
Coefficient
|
t-Stat
|
SE Reg
|
mu*
|
sig*
|
|
Obs
|
|
Pooled
|
-0.82693
|
-10.658
|
1.011
|
-0.539
|
0.860
|
|
186
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
· CREDIT
|
Null Hypothesis: Unit root (common unit root process)
|
|
|
Series: CREDIT
|
|
|
|
|
|
|
Date: 05/07/21 Time: 10:52
|
|
|
|
|
|
Sample: 1980 2019
|
|
|
|
|
|
|
Exogenous variables: None
|
|
|
|
|
|
Automatic selection of maximum lags
|
|
|
|
|
|
Automatic lag length selection based on SIC: 0 to 1
|
|
|
|
Newey-West automatic bandwidth selection and Bartlett
kernel
|
|
|
Total number of observations: 194
|
|
|
|
|
|
Cross-sections included: 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Method
|
|
|
Statistic
|
|
Prob.**
|
|
|
Levin, Lin & Chu t*
|
|
-2.53251
|
|
0.0057
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
** Probabilities are computed assuming asympotic normality
|
|
|
|
|
|
|
|
|
|
|
Intermediate results on CREDIT
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cross
|
2nd Stage
|
Variance
|
HAC of
|
|
Max
|
Band-
|
|
|
section
|
Coefficient
|
of Reg
|
Dep.
|
Lag
|
Lag
|
width
|
Obs
|
|
Gabon
|
-0.03222
|
9.6835
|
3.4157
|
0
|
9
|
13.0
|
39
|
|
Cameroun
|
-0.03852
|
6.4372
|
6.8894
|
0
|
9
|
0.0
|
39
|
|
Congo
|
-0.02175
|
12.315
|
12.370
|
0
|
9
|
7.0
|
39
|
|
Guinée
|
-0.05835
|
21.909
|
17.439
|
0
|
9
|
4.0
|
39
|
|
Tchad
|
-0.07045
|
6.6322
|
4.5011
|
1
|
9
|
9.0
|
38
|
|
|
|
|
|
|
|
|
|
Coefficient
|
t-Stat
|
SE Reg
|
mu*
|
sig*
|
|
Obs
|
|
Pooled
|
-0.04021
|
-2.587
|
1.002
|
0.002
|
1.023
|
|
194
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
· IDE
|
Null Hypothesis: Unit root (common unit root process)
|
|
|
Series: IDE
|
|
|
|
|
|
|
|
Date: 05/07/21 Time: 10:54
|
|
|
|
|
|
Sample: 1980 2019
|
|
|
|
|
|
|
Exogenous variables: Individual effects
|
|
|
|
|
Automatic selection of maximum lags
|
|
|
|
|
|
Automatic lag length selection based on SIC: 0 to 3
|
|
|
|
Newey-West automatic bandwidth selection and Bartlett
kernel
|
|
|
Total number of observations: 153
|
|
|
|
|
|
Cross-sections included: 4 (1 dropped)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Method
|
|
|
Statistic
|
|
Prob.**
|
|
|
Levin, Lin & Chu t*
|
|
-4.62558
|
|
0.0000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
** Probabilities are computed assuming asympotic normality
|
|
|
|
|
|
|
|
|
|
|
Intermediate results on IDE
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cross
|
2nd Stage
|
Variance
|
HAC of
|
|
Max
|
Band-
|
|
|
section
|
Coefficient
|
of Reg
|
Dep.
|
Lag
|
Lag
|
width
|
Obs
|
|
Gabon
|
-0.63710
|
3.3173
|
0.3921
|
0
|
9
|
24.0
|
39
|
|
Cameroun
|
-0.98044
|
0.2410
|
0.0328
|
0
|
9
|
15.0
|
39
|
|
Congo
|
-0.40228
|
9.E+16
|
6.E+16
|
0
|
9
|
3.0
|
39
|
|
Guinée
|
|
Dropped from Test
|
|
|
|
|
Tchad
|
-0.18569
|
1.E+13
|
3.E+12
|
3
|
9
|
15.0
|
36
|
|
|
|
|
|
|
|
|
|
Coefficient
|
t-Stat
|
SE Reg
|
mu*
|
sig*
|
|
Obs
|
|
Pooled
|
-0.51658
|
-6.869
|
1.049
|
-0.539
|
0.857
|
|
153
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
· INST
|
Null Hypothesis: Unit root (common unit root process)
|
|
|
Series: D(INST)
|
|
|
|
|
|
|
Date: 05/07/21 Time: 10:58
|
|
|
|
|
|
Sample: 1980 2019
|
|
|
|
|
|
|
Exogenous variables: Individual effects
|
|
|
|
|
Automatic selection of maximum lags
|
|
|
|
|
|
Automatic lag length selection based on SIC: 0 to 1
|
|
|
|
Newey-West automatic bandwidth selection and Bartlett
kernel
|
|
|
Total number of observations: 189
|
|
|
|
|
|
Cross-sections included: 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Method
|
|
|
Statistic
|
|
Prob.**
|
|
|
Levin, Lin & Chu t*
|
|
-12.6937
|
|
0.0000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
** Probabilities are computed assuming asympotic normality
|
|
|
|
|
|
|
|
|
|
|
Intermediate results on D(INST)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cross
|
2nd Stage
|
Variance
|
HAC of
|
|
Max
|
Band-
|
|
|
section
|
Coefficient
|
of Reg
|
Dep.
|
Lag
|
Lag
|
width
|
Obs
|
|
Gabon
|
-1.72059
|
1.4046
|
0.5463
|
1
|
9
|
8.0
|
37
|
|
Cameroun
|
-1.20066
|
0.8513
|
0.0562
|
0
|
9
|
33.0
|
38
|
|
Congo
|
-1.07762
|
1.2665
|
0.2299
|
0
|
9
|
14.0
|
38
|
|
Guinée
|
-1.21948
|
5.1342
|
1.9025
|
0
|
9
|
5.0
|
38
|
|
Tchad
|
-0.69288
|
1.1021
|
0.1322
|
0
|
9
|
18.0
|
38
|
|
|
|
|
|
|
|
|
|
Coefficient
|
t-Stat
|
SE Reg
|
mu*
|
sig*
|
|
Obs
|
|
Pooled
|
-1.11058
|
-14.196
|
1.038
|
-0.539
|
0.860
|
|
189
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
· XCE
|
Null Hypothesis: Unit root (common unit root process)
|
|
|
Series: XCE
|
|
|
|
|
|
|
|
Date: 05/07/21 Time: 11:01
|
|
|
|
|
|
Sample: 1980 2019
|
|
|
|
|
|
|
Exogenous variables: Individual effects
|
|
|
|
|
Automatic selection of maximum lags
|
|
|
|
|
|
Automatic lag length selection based on SIC: 0 to 1
|
|
|
|
Newey-West automatic bandwidth selection and Bartlett
kernel
|
|
|
Total number of observations: 194
|
|
|
|
|
|
Cross-sections included: 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Method
|
|
|
Statistic
|
|
Prob.**
|
|
|
Levin, Lin & Chu t*
|
|
-6.00944
|
|
0.0000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
** Probabilities are computed assuming asympotic normality
|
|
|
|
|
|
|
|
|
|
|
Intermediate results on XCE
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cross
|
2nd Stage
|
Variance
|
HAC of
|
|
Max
|
Band-
|
|
|
section
|
Coefficient
|
of Reg
|
Dep.
|
Lag
|
Lag
|
width
|
Obs
|
|
Gabon
|
-1.06137
|
23.899
|
1.2431
|
0
|
9
|
38.0
|
39
|
|
Cameroun
|
-0.31265
|
7.5825
|
11.107
|
1
|
9
|
2.0
|
38
|
|
Congo
|
-0.48359
|
18.914
|
7.7212
|
0
|
9
|
13.0
|
39
|
|
Guinée
|
-0.55266
|
580.26
|
110.53
|
0
|
9
|
15.0
|
39
|
|
Tchad
|
-0.91377
|
65.295
|
69.502
|
0
|
9
|
1.0
|
39
|
|
|
|
|
|
|
|
|
|
Coefficient
|
t-Stat
|
SE Reg
|
mu*
|
sig*
|
|
Obs
|
|
Pooled
|
-0.59345
|
-9.334
|
1.050
|
-0.539
|
0.857
|
|
194
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
· CS
|
Null Hypothesis: Unit root (common unit root process)
|
|
|
Series: D(CS)
|
|
|
|
|
|
|
Date: 05/07/21 Time: 11:00
|
|
|
|
|
|
Sample: 1980 2019
|
|
|
|
|
|
|
Exogenous variables: Individual effects
|
|
|
|
|
Automatic selection of maximum lags
|
|
|
|
|
|
Automatic lag length selection based on SIC: 0 to 6
|
|
|
|
Newey-West automatic bandwidth selection and Bartlett
kernel
|
|
|
Total number of observations: 184
|
|
|
|
|
|
Cross-sections included: 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Method
|
|
|
Statistic
|
|
Prob.**
|
|
|
Levin, Lin & Chu t*
|
|
-10.7495
|
|
0.0000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
** Probabilities are computed assuming asympotic normality
|
|
|
|
|
|
|
|
|
|
|
Intermediate results on D(CS)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cross
|
2nd Stage
|
Variance
|
HAC of
|
|
Max
|
Band-
|
|
|
section
|
Coefficient
|
of Reg
|
Dep.
|
Lag
|
Lag
|
width
|
Obs
|
|
Gabon
|
-1.24068
|
33.401
|
8.1567
|
0
|
9
|
8.0
|
38
|
|
Cameroun
|
-0.97143
|
2.9765
|
1.1057
|
0
|
9
|
5.0
|
38
|
|
Congo
|
-1.06587
|
45.232
|
15.913
|
0
|
9
|
6.0
|
38
|
|
Guinée
|
1.47006
|
3.5352
|
0.8693
|
6
|
9
|
37.0
|
32
|
|
Tchad
|
-1.32756
|
47.401
|
7.2088
|
0
|
9
|
14.0
|
38
|
|
|
|
|
|
|
|
|
|
Coefficient
|
t-Stat
|
SE Reg
|
mu*
|
sig*
|
|
Obs
|
|
Pooled
|
-1.07576
|
-12.809
|
1.094
|
-0.540
|
0.864
|
|
184
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Annexe 3 : Test de Pedroni
|
Pedroni Residual Cointegration Test
|
|
|
|
Series: XCE CS DINST DINST_CS IDE INF EXP01
|
|
|
Date: 05/07/21 Time: 22:36
|
|
|
|
Sample: 1980 2019
|
|
|
|
|
Included observations: 200
|
|
|
|
Cross-sections included: 5
|
|
|
|
Null Hypothesis: No cointegration
|
|
|
|
Trend assumption: No deterministic trend
|
|
|
Automatic lag length selection based on SIC with a max lag of
8
|
|
Newey-West automatic bandwidth selection and Bartlett
kernel
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Alternative hypothesis: common AR coefs. (within-dimension)
|
|
|
|
|
Weighted
|
|
|
|
Statistic
|
Prob.
|
Statistic
|
Prob.
|
|
Panel v-Statistic
|
-2.483187
|
0.9935
|
-2.480167
|
0.9934
|
|
Panel rho-Statistic
|
-0.599359
|
0.2745
|
-1.027589
|
0.1521
|
|
Panel PP-Statistic
|
-3.052106
|
0.0011
|
-4.803139
|
0.0000
|
|
Panel ADF-Statistic
|
-3.079623
|
0.0010
|
-4.907662
|
0.0000
|
|
|
|
|
|
|
|
Alternative hypothesis: individual AR coefs.
(between-dimension)
|
|
|
|
|
|
|
|
|
Statistic
|
Prob.
|
|
|
|
Group rho-Statistic
|
-0.975188
|
0.1647
|
|
|
|
Group PP-Statistic
|
-6.202879
|
0.0000
|
|
|
|
Group ADF-Statistic
|
-6.126803
|
0.0000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cross section specific results
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Phillips-Peron results (non-parametric)
|
|
|
|
|
|
|
|
|
Cross ID
|
AR(1)
|
Variance
|
HAC
|
Bandwidth
|
Obs
|
|
Gabon
|
0.101
|
21.52081
|
15.81998
|
6.00
|
38
|
|
Cameroun
|
0.182
|
6.936069
|
9.324657
|
2.00
|
38
|
|
Congo
|
0.174
|
16.74248
|
8.000443
|
12.00
|
38
|
|
Guinée
|
0.310
|
535.1802
|
525.1947
|
2.00
|
38
|
|
Tchad
|
-0.132
|
37.11825
|
48.78840
|
4.00
|
38
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller results (parametric)
|
|
|
|
|
|
|
|
|
Cross ID
|
AR(1)
|
Variance
|
Lag
|
Max lag
|
Obs
|
|
Gabon
|
0.101
|
21.52081
|
0
|
8
|
38
|
|
Cameroun
|
0.182
|
6.936069
|
0
|
8
|
38
|
|
Congo
|
0.174
|
16.74248
|
0
|
8
|
38
|
|
Guinée
|
0.310
|
535.1802
|
0
|
8
|
38
|
|
Tchad
|
-0.132
|
37.11825
|
0
|
8
|
38
|
|
|
|
|
|
|
|
|
|
|
|
|
Annexe 4 : Résultat de
l'estimation GMM
|
Dependent Variable: XCE
|
|
|
|
Method: Panel Generalized Method of Moments
|
|
|
Transformation: Orthogonal Deviations
|
|
|
Date: 05/07/21 Time: 23:32
|
|
|
|
Sample (adjusted): 1982 2019
|
|
|
|
Periods included: 38
|
|
|
|
Cross-sections included: 5
|
|
|
|
Total panel (balanced) observations: 190
|
|
|
2SLS instrument weighting matrix
|
|
|
Instrument specification: XCE DCS DINST DINST_CS IDE INF
CREDIT
|
|
EXP01
|
|
|
|
|
Constant added to instrument list
|
|
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
|
XCE(-1)
|
2.650750
|
0.489303
|
5.417404
|
0.0000
|
|
DCS
|
0.426088
|
2.491834
|
-0.170994
|
0,0219
|
|
DINST
|
0.662626
|
3.418802
|
0.193818
|
0,0429
|
|
DINST_CS
|
-0.010777
|
0.049399
|
0.218168
|
0,0645
|
|
IDE
|
1.27E-08
|
2.02E-08
|
0.627221
|
0.0301
|
|
INF
|
0.233399
|
0.215444
|
1.083342
|
0.5313
|
|
CREDIT
|
0.645886
|
0.377499
|
1.710960
|
0.0888
|
|
EXP01
|
-4.72E-11
|
4.63E-11
|
-1.020269
|
0.3090
|
|
|
|
|
|
|
|
|
|
|
|
Effects Specification
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cross-section fixed (orthogonal deviations)
|
|
|
|
|
|
|
|
|
|
|
|
|
Mean dependent var
|
1.675434
|
S.D. dependent var
|
13.22506
|
|
S.E. of regression
|
32.26578
|
Sum squared resid
|
189476.6
|
|
J-statistic
|
0,230140
|
Instrument rank
|
8
|
|
|
|
|
|
|
|
|
|
|
Annexe 5 : Test de
sur-identification
|
Arellano-Bond Serial Correlation Test
|
|
|
Equation: Untitled
|
|
|
|
Date: 05/08/21 Time: 00:06
|
|
|
|
Sample: 1980 2019
|
|
|
|
Included observations: 190
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Test order
|
m-Statistic
|
rho
|
SE(rho)
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
|
AR(1)
|
-1.328508
|
-20134.069825
|
15155.394728
|
0.0020
|
|
AR(2)
|
-3.092682
|
-61090.051247
|
19753.099202
|
0.1840
|
|
|
|
|
|
|
|
|
|
|
| 
