A Niveau
Null Hypothesis: LDFONCT has a unit root
|
Exogenous: None
|
|
|
Lag Length: 0 (Automatic - based on SIC, maxlag=7)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
t-Statistic
|
Prob.*
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller test statistic
|
-0.193296
|
0.6082
|
Test critical values:
|
1% level
|
|
-2.644302
|
|
|
5% level
|
|
-1.952473
|
|
|
10% level
|
|
-1.610211
|
|
|
|
|
|
|
|
|
|
|
|
*MacKinnon (1996) one-sided p-values.
|
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller Test Equation
|
|
Dependent Variable: D(LDFONCT)
|
|
Method: Least Squares
|
|
|
Date: 07/26/16 Time: 15:04
|
|
Sample (adjusted): 1986 2015
|
|
Included observations: 30 afteradjustments
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
LDFONCT(-1)
|
-0.000545
|
0.002820
|
-0.193296
|
0.8481
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Après différence
1ère
Null Hypothesis: DLDFONCT has a unit root
|
Exogenous: None
|
|
|
Lag Length: 0 (Automatic - based on SIC, maxlag=7)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
t-Statistic
|
Prob.*
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller test statistic
|
-7.631884
|
0.0000
|
Test critical values:
|
1% level
|
|
-2.647120
|
|
|
5% level
|
|
-1.952910
|
|
|
10% level
|
|
-1.610011
|
|
|
|
|
|
|
|
|
|
|
|
*MacKinnon (1996) one-sided p-values.
|
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller Test Equation
|
|
Dependent Variable: D(DLDFONCT)
|
|
Method: Least Squares
|
|
|
Date: 07/26/16 Time: 15:10
|
|
Sample (adjusted): 1987 2015
|
|
Included observations: 29 afteradjustments
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
DLDFONCT(-1)
|
-1.350792
|
0.176993
|
-7.631884
|
0.0000
|
c) Log Dépenses d'Investissement :
Linv
A Niveau
Null Hypothesis: LINV has a unit root
|
|
Exogenous: Constant
|
|
|
Lag Length: 0 (Automatic - based on SIC, maxlag=7)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
t-Statistic
|
Prob.*
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller test statistic
|
-3.570594
|
0.0127
|
Test critical values:
|
1% level
|
|
-3.670170
|
|
|
5% level
|
|
-2.963972
|
|
|
10% level
|
|
-2.621007
|
|
|
|
|
|
|
|
|
|
|
|
*MacKinnon (1996) one-sided p-values.
|
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller Test Equation
|
|
Dependent Variable: D(LINV)
|
|
Method: Least Squares
|
|
|
Date: 07/26/16 Time: 15:13
|
|
Sample (adjusted): 1986 2015
|
|
Included observations: 30 afteradjustments
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
LINV(-1)
|
-0.622867
|
0.174444
|
-3.570594
|
0.0013
|
C
|
1.393495
|
0.407574
|
3.419000
|
0.0019
|
|
|
|
|
|
|
|
|
|
|
d) Log Corruption
A Niveau
Null Hypothesis: LCORR has a unit root
|
|
Exogenous: Constant, Linear Trend
|
|
Lag Length: 0 (Automatic - based on SIC, maxlag=7)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
t-Statistic
|
Prob.*
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller test statistic
|
-3.132599
|
0.1173
|
Test critical values:
|
1% level
|
|
-4.296729
|
|
|
5% level
|
|
-3.568379
|
|
|
10% level
|
|
-3.218382
|
|
|
|
|
|
|
|
|
|
|
|
*MacKinnon (1996) one-sided p-values.
|
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller Test Equation
|
|
Dependent Variable: D(LCORR)
|
|
Method: Least Squares
|
|
|
Date: 08/01/16 Time: 12:27
|
|
Sample (adjusted): 1986 2015
|
|
Included observations: 30 afteradjustments
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
LCORR(-1)
|
-0.542481
|
0.173173
|
-3.132599
|
0.0041
|
C
|
1.265978
|
0.408228
|
3.101156
|
0.0045
|
@TREND(1985)
|
-0.006300
|
0.002434
|
-2.588454
|
0.0153
|
|
|
|
|
|
|
|
|
|
|
Après Différence
1ère
Null Hypothesis: DLCORR has a unit root
|
|
Exogenous: None
|
|
|
Lag Length: 1 (Automatic - based on SIC, maxlag=7)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
t-Statistic
|
Prob.*
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller test statistic
|
-6.509666
|
0.0000
|
Test critical values:
|
1% level
|
|
-2.650145
|
|
|
5% level
|
|
-1.953381
|
|
|
10% level
|
|
-1.609798
|
|
|
|
|
|
|
|
|
|
|
|
*MacKinnon (1996) one-sided p-values.
|
|
|
|
|
|
|
|
|
|
|
|
Augmented Dickey-Fuller Test Equation
|
|
Dependent Variable: D(DLCORR)
|
|
Method: Least Squares
|
|
|
Date: 08/01/16 Time: 12:34
|
|
Sample (adjusted): 1988 2015
|
|
Included observations: 28 afteradjustments
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
DLCORR(-1)
|
-1.549590
|
0.238045
|
-6.509666
|
0.0000
|
D(DLCORR(-1))
|
0.526313
|
0.167030
|
3.151017
|
0.0041
|
|
|
|
|
|
|
|
|
|
|
3. ESTIMATION PAR MCO
Modèle I
Dependent Variable: LTXCR
|
|
|
Method: Least Squares
|
|
|
Date: 10/29/16 Time: 17:29
|
|
Sample (adjusted): 1987 2015
|
|
Included observations: 15 afteradjustments
|
Convergence achievedafter 52 iterations
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
C
|
0.438497
|
1.905486
|
0.230124
|
0.8231
|
DLDEXT
|
-0.156192
|
0.513394
|
-0.304234
|
0.7679
|
DLDFONCT
|
-0.050589
|
0.589856
|
-0.085765
|
0.9335
|
LINV
|
0.581122
|
0.737898
|
0.787537
|
0.4512
|
DLCORR
|
-1.809516
|
1.144973
|
-1.580401
|
0.1485
|
AR(1)
|
1.546183
|
0.578481
|
2.672834
|
0.0255
|
|
|
|
|
|
|
|
|
|
|
R-squared
|
0.458704
|
Meandependent var
|
1.611605
|
Adjusted R-squared
|
0.157984
|
S.D. dependent var
|
0.737553
|
S.E. of regression
|
0.676790
|
Akaike info criterion
|
2.346262
|
Sumsquaredresid
|
4.122397
|
Schwarz criterion
|
2.629482
|
Log likelihood
|
-11.59696
|
Hannan-Quinn criter.
|
2.343245
|
F-statistic
|
1.525351
|
Durbin-Watson stat
|
1.621474
|
Prob(F-statistic)
|
0.273934
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Modèle II
Dependent Variable: LTXCR
|
|
|
Method: Least Squares
|
|
|
Date: 10/29/16 Time: 17:43
|
|
Sample (adjusted): 1987 2015
|
|
Included observations: 15 afteradjustments
|
Convergence achievedafter 109 iterations
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
C
|
-971.0400
|
5082072.
|
-0.000191
|
0.9999
|
DLDEXT
|
-0.104211
|
0.620297
|
-0.168002
|
0.8699
|
LINV
|
0.166882
|
1.074690
|
0.155284
|
0.8797
|
DLCORR
|
-1.783192
|
1.478344
|
-1.206209
|
0.2555
|
AR(1)
|
0.999894
|
0.555995
|
1.798388
|
0.1023
|
|
|
|
|
|
|
|
|
|
|
R-squared
|
0.410395
|
Meandependent var
|
1.611605
|
Adjusted R-squared
|
0.174553
|
S.D. dependent var
|
0.737553
|
S.E. of regression
|
0.670098
|
Akaike info criterion
|
2.298415
|
Sumsquaredresid
|
4.490307
|
Schwarz criterion
|
2.534431
|
Log likelihood
|
-12.23811
|
Hannan-Quinn criter.
|
2.295900
|
F-statistic
|
1.740127
|
Durbin-Watson stat
|
0.989953
|
Prob(F-statistic)
|
0.217464
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Modèle III
Dependent Variable: LTXCR
|
|
|
Method: Least Squares
|
|
|
Date: 10/31/16 Time: 17:25
|
|
Sample (adjusted): 1987 2015
|
|
Included observations: 15 afteradjustments
|
Convergence achievedafter 10 iterations
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
C
|
2.030455
|
0.779553
|
2.604641
|
0.0245
|
DLDEXT
|
-0.110993
|
0.497730
|
-0.222999
|
0.8276
|
DLCORR
|
-1.740160
|
1.190569
|
-1.461620
|
0.1718
|
AR(1)
|
1.297455
|
0.496899
|
2.611104
|
0.0242
|
|
|
|
|
|
|
|
|
|
|
R-squared
|
0.428672
|
Meandependent var
|
1.611605
|
Adjusted R-squared
|
0.272855
|
S.D. dependent var
|
0.737553
|
S.E. of regression
|
0.628932
|
Akaike info criterion
|
2.133592
|
Sumsquaredresid
|
4.351113
|
Schwarz criterion
|
2.322405
|
Log likelihood
|
-12.00194
|
Hannan-Quinn criter.
|
2.131581
|
F-statistic
|
2.751129
|
Durbin-Watson stat
|
1.125338
|
Prob(F-statistic)
|
0.093095
|
|
|
|
Modèle IV
Dependent Variable: LTXCR
|
|
|
Method: Least Squares
|
|
|
Date: 10/31/16 Time: 18:20
|
|
Sample (adjusted): 1987 2015
|
|
Included observations: 15 afteradjustments
|
Convergence achievedafter 4 iterations
|
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
C
|
2.047845
|
0.777087
|
2.635285
|
0.0218
|
DLCORR
|
-1.800103
|
1.120816
|
-1.606065
|
0.1342
|
AR(1)
|
1.284825
|
0.454245
|
2.828486
|
0.0152
|
|
|
|
|
|
|
|
|
|
|
R-squared
|
0.425967
|
Meandependent var
|
1.611605
|
Adjusted R-squared
|
0.330295
|
S.D. dependent var
|
0.737553
|
S.E. of regression
|
0.603581
|
Akaike info criterion
|
2.004982
|
Sumsquaredresid
|
4.371714
|
Schwarz criterion
|
2.146592
|
Log likelihood
|
-12.03737
|
Hannan-Quinn criter.
|
2.003474
|
F-statistic
|
4.452358
|
Durbin-Watson stat
|
1.234667
|
Prob(F-statistic)
|
0.035778
|
|
|
|
4. ESTIMATION PAR GMM
Modèle I
Dependent Variable: LTXCR
|
|
|
Method: Generalized Method of Moments
|
|
Date: 07/30/16 Time: 20:11
|
|
Sample (adjusted): 1987 2015
|
|
Included observations: 15 afteradjustments
|
Sequential 1-step weighting matrix & coefficient iteration
|
Estimation weighting matrix: HAC (Bartlett kernel, Newey-West
fixed
|
bandwidth =
3.0000)
|
|
Standard errors & covariance computed using estimation
weighting
|
Matrix
|
|
|
Convergence achievedafter 17 iterations
|
Instrument specification: DLDEXT DLDFONCT LINV DLCORR
|
Constant added to instrument list
|
|
Lagged dependent variable ®ressors added to instrument
list
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
DLDEXT
|
-0.044811
|
0.130383
|
-0.343690
|
0.7382
|
DLDFONCT
|
0.017087
|
0.124112
|
0.137676
|
0.8932
|
LINV
|
0.793033
|
0.087713
|
9.041265
|
0.0000
|
DLCORR
|
-1.531804
|
0.712394
|
-2.150221
|
0.0570
|
AR(1)
|
1.320561
|
0.356069
|
3.708725
|
0.0040
|
|
|
|
|
|
|
|
|
|
|
R-squared
|
0.434905
|
Meandependent var
|
1.611605
|
Adjusted R-squared
|
0.208868
|
S.D. dependent var
|
0.737553
|
S.E. of regression
|
0.656021
|
Sumsquaredresid
|
4.303641
|
Durbin-Watson stat
|
1.612781
|
J-statistic
|
3.570737
|
Instrument rank
|
10
|
Prob(J-statistic)
|
0.612712
|
|
|
|
|
|
|
|
|
|
|
Modèle II
5. TESTS STATISTIQUES
Test de Normalité/ Modèle IV
estimé par MCO

Test de Normalité/ Modèle III
estimé par GMM

Test d'Homoscédasticité (Test de White)/
Modèle IV estimé par MCO
Heteroskedasticity Test: White
|
|
|
|
|
|
|
|
|
|
|
|
F-statistic
|
40.04501
|
Prob. F(2,12)
|
0.0000
|
Obs*R-squared
|
13.04539
|
Prob. Chi-Square(2)
|
0.0015
|
Scaledexplained SS
|
13.33007
|
Prob. Chi-Square(2)
|
0.0013
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Test Equation:
|
|
|
Dependent Variable: RESID^2
|
|
Method: Least Squares
|
|
|
Date: 11/07/16 Time: 20:18
|
|
Sample: 1987 2015
|
|
|
Included observations: 15
|
|
|
Collinear test regressors dropped from specification
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
|
|
|
|
|
|
|
|
|
|
C
|
-0.029494
|
0.076209
|
-0.387019
|
0.7055
|
GRADF_02^2
|
-1.153875
|
1.926733
|
-0.598876
|
0.5604
|
GRADF_03^2
|
1.445558
|
0.161877
|
8.929975
|
0.0000
|
|
|
|
|
|
|
|
|
|
|
R-squared
|
0.869693
|
Meandependent var
|
0.291448
|
Adjusted R-squared
|
0.847975
|
S.D. dependent var
|
0.539082
|
S.E. of regression
|
0.210190
|
Akaike info criterion
|
-0.104751
|
Sumsquaredresid
|
0.530159
|
Schwarz criterion
|
0.036859
|
Log likelihood
|
3.785634
|
Hannan-Quinn criter.
|
-0.106260
|
F-statistic
|
40.04501
|
Durbin-Watson stat
|
1.267808
|
Prob(F-statistic)
|
0.000005
|
|
|
|
Test de Stabilité (Test de Cusum)/
Modèle IV estimé par MCO

|