Conclusion : M3_M1 est I(1)
Etude de la série LNM3M1 :
LNM3_M1
|
13.0 12.8 12.6 12.4 12.2 12.0 11.8 11.6
|
|
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Le graphe de cette série se présente comme suit
:
55
Estimation du modèle [3] pour LNM3_M1
:
|
Null Hypothesis: LNM3_M1 has a unit root Exogenous: Constant,
Linear Trend
Lag Length: 0 (Automatic - based on SIC, maxlag=9)
|
|
|
t-Statistic
|
Prob.*
|
|
Augmented Dickey-Fuller test statistic -1.474968
|
0.8212
|
|
Test critical values: 1% level -4.211868
|
|
|
5% level -3.529758
|
|
|
10% level -3.196411
|
|
|
*MacKinnon (1996) one-sided p-values.
|
|
|
Augmented Dickey-Fuller Test Equation
|
|
|
Dependent Variable: D(LNM3_M1)
|
|
|
Method: Least Squares
|
|
|
Date: 06/12/13 Time: 09:31
|
|
|
Sample (adjusted): 2002Q2 2011Q4
|
|
|
Included observations: 39 after adjustments
|
|
|
Variable Coefficient Std. Error t-Statistic
|
Prob.
|
|
LNM3_M1(-1) -0.108287 0.073416 -1.474968
|
0.1489
|
|
C 1.287443 0.855440 1.505007
|
0.1410
|
|
@TREND("2002Q1") 0.003483 0.002377 1.465434
|
0.1515
|
|
R-squared 0.057134 Mean dependent var
|
0.026420
|
|
Adjusted R-squared 0.004752 S.D. dependent var
|
0.029508
|
|
S.E. of regression 0.029438 Akaike info criterion
|
-4.139289
|
|
Sum squared resid 0.031196 Schwarz criterion
|
-4.011323
|
|
Log likelihood 83.71613 Hannan-Quinn criter.
|
-4.093376
|
|
F-statistic 1.090721 Durbin-Watson stat
|
1.505514
|
|
Prob(F-statistic) 0.346819
|
|
|
|
La statistique relative à Ö1 (-1.47) est
supérieure à sa valeur critique (-3.53). On accepte donc H0, et
on passe au test de l'hypothèse H0,3 :
|
Wald Test: Equation: Untitled
|
|
|
|
Test Statistic Value
|
df
|
Probability
|
|
F-statistic 1.090721
Chi-square 2.181442
|
(2, 36)
2
|
0.3468
0.3360
|
|
Null Hypothesis: C(1)=0,C(3)=0 Null Hypothesis Summary:
|
|
|
|
Normalized Restriction (= 0)
|
Value
|
Std. Err.
|
|
C(1) C(3)
|
-0.108287
0.003483
|
0.073416
0.002377
|
|
Restrictions are linear in coefficients.
|
|
La p-value relative à F étant supérieure
à 5%, on accepte H0,3 et on passe à l'estimation du
modèle [2].
56
Estimation du modèle [2] pour LNM3_M1
:
|
Null Hypothesis: LNM3_M1 has a unit root Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=9)
|
|
|
t-Statistic
|
Prob.*
|
|
Augmented Dickey-Fuller test statistic -0.181452
|
0.9325
|
|
Test critical values: 1% level -3.610453
5% level -2.938987
10% level -2.607932
|
|
|
*MacKinnon (1996) one-sided p-values.
Augmented Dickey-Fuller Test Equation Dependent Variable:
D(LNM3_M1) Method: Least Squares
Date: 06/12/13 Time: 09:36
Sample (adjusted): 2002Q2 2011Q4 Included observations: 39 after
adjustments
|
|
|
Variable Coefficient Std. Error t-Statistic
|
Prob.
|
|
LNM3_M1(-1) -0.002384 0.013137 -0.181452
C 0.055712 0.161502 0.344962
|
0.8570
0.7321
|
|
R-squared 0.000889 Mean dependent var
Adjusted R-squared -0.026114 S.D. dependent var
S.E. of regression 0.029891 Akaike info criterion
Sum squared resid 0.033057 Schwarz criterion
Log likelihood 82.58628 Hannan-Quinn criter.
F-statistic 0.032925 Durbin-Watson stat
Prob(F-statistic) 0.857003
|
0.026420 0.029508 -4.132630 -4.047319 -4.102021 1.582537
|
|
|
La statistique relative à Ö1 (-0.18) est
supérieure à sa valeur critique (-2.94). On accepte donc H0, et
on passe au test de l'hypothèse H0,2 :
|
Wald Test: Equation: Untitled
|
|
|
|
Test Statistic Value
|
df
|
Probability
|
|
F-statistic 15.25127
Chi-square 30.50254
|
(2, 37)
2
|
0.0000
0.0000
|
|
Null Hypothesis: C(1)=0,C(2)=0 Null Hypothesis Summary:
|
|
|
|
Normalized Restriction (= 0)
|
Value
|
Std. Err.
|
C(1)
C(2)
|
-0.002384
0.055712
|
0.013137
0.161502
|
|
Restrictions are linear in coefficients.
|
|
La p-value relative à F étant inférieure
à 5%, on rejette H0,2 et on conclut que LNM3_M1 est DS avec
Drift. La meilleure façon de la stationnariser est de la
différencier. Le graphe de cette nouvelle série se
présente comme suit :
LNM3_M1D1
|
.10 .08 .06 .04 .02 .00 -.02 -.04
|
|
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
57
Les tests ADF conduits sur cette série montrent qu'elle
est stationnaire autour d'une moyenne.
| 
