Analyse de l'investissement en infrastructure routière sur la croissance économique au Bénin

Chapitre 2 Analyse des résultats ; implication économique et recommandations

30

Paragraphe 1 : Présentation et analyse des résultats du modèle à correction d'erreur (ECM) 30

A-Présentation des résultats du modèle à correction d'erreur (ECM) 30

B-Validation et analyse des résultats du modèle à correction d'erreur 31

41

Paragraphe 2 : Recommandations de politique économique 34

CONCLUSION GENERALE 37

ANNEXES 41

ANNEXES
ANNEXE1

TableauA1.1: Résultat du test d'ADF sur lmhat en niveau

Null Hypothesis: LMHAT has a unit root

Exogenous: Constant, Linear Trend

Lag Length: 10 (Automatic based on AIC, MAXLAG=10)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -6.397118 0.0007

Test critical values: 1% level -4.728363

5% level -3.759743

10% level -3.324976

*MacKinnon (1996) one-sided p-values.

Warning: Probabilities and critical values calculated for 20 observations and may not be accurate for a sample size of 15

Augmented Dickey-Fuller Test Equation Dependent Variable: D(LMHAT)

Method: Least Squares

Date: 03/17/11 Time: 06:03

Sample (adjusted): 1996 2010

Included observations: 15 after adjustments

Variable	Coefficient	Std. Error t-Statistic	Prob.
LMHAT(-1)	-1.112236	0.173865 -6.397118	0.0236
D(LMHAT(-1))	0.149154	0.088156 1.691944	0.2327
D(LMHAT(-2))	-0.014857	0.008149 -1.823247	0.2098
D(LMHAT(-3))	-0.041041	0.006962 -5.894976	0.0276
D(LMHAT(-4))	-0.016390	0.007356 -2.228109	0.1557
D(LMHAT(-5))	0.002288	0.005453 -0.419681	0.7155
D(LMHAT(-6))	-0.002308	0.000341 -6.760197	0.0212
D(LMHAT(-7))	0.001985	0.000382 5.196723	0.0351
D(LMHAT(-8))	0.000144	0.000338 0.427326	0.7108
D(LMHAT(-9))	0.000355	0.000271 1.309987	0.3204
D(LMHAT(-10))	0.000284	7.15E-05 3.970410	0.0580
C	24.35526	3.803071 6.404102	0.0235
@TREND(1985)	0.022019	0.003442 6.397660	0.0236
R-squared	1.000000	Mean dependent var	0.020124
Adjusted R-squared	0.999998	S.D. dependent var	0.000623
S.E. of regression	8.85E-07	Akaike info criterion	-25.31924
Sum squared resid	1.57E-12	Schwarz criterion	-24.70560
Log likelihood	202.8943	Hannan-Quinn criter.	-25.32578

42

F-statistic 577608.2 Durbin-Watson stat 2.565604

Prob(F-statistic) 0.000002

TableauA1.2: Resultat du test d'ADF sur lmhat en difference premiere

Null Hypothesis: D(LMHAT) has a unit root

Exogenous: Constant

Lag Length: 10 (Automatic based on AIC, MAXLAG=10)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -5.356692 0.0010

Test critical values: 1% level -4.004425

5% level -3.098896

10% level -2.690439

*MacKinnon (1996) one-sided p-values.

Warning: Probabilities and critical values calculated for 20 observations and may not be accurate for a sample size of 14

Augmented Dickey-Fuller Test Equation Dependent Variable: D(LMHAT,2)

Method: Least Squares

Date: 03/17/11 Time: 06:09

Sample (adjusted): 1997 2010

Included observations: 14 after adjustments

Variable	Coefficient	Std. Error t-Statistic	Prob.
D(LMHAT(-1))	-1.189197	0.222002 -5.356692	0.0331
D(LMHAT(-1),2)	0.178938	0.090248 1.982749	0.1859
D(LMHAT(-2),2)	-0.015002	0.007431 -2.018849	0.1810
D(LMHAT(-3),2)	-0.039403	0.006437 -6.121366	0.0257
D(LMHAT(-4),2)	-0.019560	0.009499 -2.059224	0.1757
D(LMHAT(-5),2)	-0.004340	0.005942 -0.730382	0.5411
D(LMHAT(-6),2)	-0.002314	0.000295 -7.834707	0.0159
D(LMHAT(-7),2)	0.001953	0.000388 5.034346	0.0373
D(LMHAT(-8),2)	0.000116	0.000317 0.364525	0.7504
D(LMHAT(-9),2)	0.000463	0.000311 1.489167	0.2749
D(LMHAT(-10),2)	0.000301	7.18E-05 4.199137	0.0523
C	0.023543	0.004395 5.357161	0.0331
R-squared	0.999999	Mean dependent var	-9.39E-05
Adjusted R-squared	0.999991	S.D. dependent var	0.000462
S.E. of regression	1.37E-06	Akaike info criterion	-24.39965
Sum squared resid	3.74E-12	Schwarz criterion	-23.85188
Log likelihood	182.7975	Hannan-Quinn criter.	-24.45035
F-statistic	135125.0	Durbin-Watson stat	2.545576
Prob(F-statistic)	0.000007

ANN EX E2

TableauA2.1: Resultat du test d'ADF sur ldep en niveau Null Hypothesis: LDEP has a unit root

43

Exogenous: None

Lag Length: 1 (Automatic based on AIC, MAXLAG=10)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic 1.493290 0.9624

Test critical values: 1% level -2.664853

5% level -1.955681

10% level -1.608793

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation Dependent Variable: D(LDEP)

Method: Least Squares

Date: 03/17/11 Time: 06:13

Sample (adjusted): 1987 2010

Included observations: 24 after adjustments

Variable	Coefficient	Std. Error t-Statistic	Prob.
LDEP(-1)	0.004733	0.003170 1.493290	0.1496
D(LDEP(-1))	-0.280629	0.204254 -1.373920	0.1833
R-squared	0.076619	Mean dependent var	0.088488
Adjusted R-squared	0.034647	S.D. dependent var	0.363935
S.E. of regression	0.357575	Akaike info criterion	0.860712
Sum squared resid	2.812918	Schwarz criterion	0.958883
Log likelihood	-8.328545	Hannan-Quinn criter.	0.886757
Durbin-Watson stat	2.063556

TableauA2.2: Résultat du test d'ADF sur ldep en difference première

Null Hypothesis: D(LDEP) has a unit root

Exogenous: None

Lag Length: 0 (Automatic based on AIC, MAXLAG=10)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -5.933345 0.0000

Test critical values: 1% level -2.664853

5% level -1.955681

10% level -1.608793

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation Dependent Variable: D(LDEP,2)

Method: Least Squares

Date: 03/17/11 Time: 06:17

Sample (adjusted): 1987 2010

Included observations: 24 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(LDEP(-1)) -1.204070 0.202933 -5.933345 0.0000

44

R-squared

	0.604838	Mean dependent var	-0.001885
Adjusted R-squared	0.604838	S.D. dependent var	0.583837
S.E. of regression	0.367011	Akaike info criterion	0.873924
Sum squared resid	3.098034	Schwarz criterion	0.923010
Log likelihood	-9.487092	Hannan-Quinn criter.	0.886947
Durbin-Watson stat	1.977388