ANNEXE 2 : Les différents tests pour le
modèle à
effets fixes : cas du maïs
|
. xtreg lprod lsup lpoprur lprix lhautpl, fe
Fixed-effects (within) regression Group variable:
comm
|
Number of obs =
Number of groups =
|
88
8
|
|
R-sq:
|
within
|
= 0.5771
|
|
|
Obs per group: min =
|
11
|
|
between
|
= 0.9206
|
|
|
avg =
|
11.0
|
|
overall
|
= 0.8790
|
|
|
max =
|
11
|
|
|
|
|
|
F(4,76) =
|
25.93
|
|
corr(u_i, Xb)
|
= 0.7783
|
|
|
Prob > F =
|
0.0000
|
|
lprod |
|
Coef.
|
Std. Err.
|
t
|
P>|t| [95% Conf.
|
Interval]
|
|
+
|
|
|
|
|
|
|
lsup |
|
.399503
|
.0719143
|
5.56
|
0.000 .2562732
|
.5427328
|
|
lpoprur |
|
.5146713
|
.1144338
|
4.50
|
0.000 .2867567
|
.7425858
|
|
lprix |
|
-.2096241
|
.128581
|
-1.63
|
0.107 -.4657153
|
.0464672
|
|
lhautpl |
|
-.1152291
|
.1195592
|
-0.96
|
0.338 -.353352
|
.1228937
|
|
|
_cons
|
.272134
|
1.524566
|
0.18
|
0.859 -2.764303
|
3.308571
|
|
+
|
|
|
|
|
|
|
sigma_u |
|
.77191572
|
|
|
|
|
|
sigma_e |
|
.26264837
|
|
|
|
|
|
rho |
|
.89623914
|
(fraction of variance due to u_i)
|
|
|
F test that all u_i=0: F(7, 76) =
|
33.58
|
Prob > F =
|
0.0000
|
|
. est store fixed
. xtreg lprod lsup lpoprur lprix lhautpl, Random-effects
GLS regression
|
re
|
Number of obs =
|
88
|
|
Group variable: comm
|
|
Number of groups =
|
8
|
|
R-sq: within = 0.5770
|
|
Obs per group: min =
|
11
|
|
between = 0.9211
|
|
avg =
|
11.0
|
|
overall = 0.8795
|
|
max =
|
11
|
|
Random effects u_i ~ Gaussian
|
|
Wald chi2(4) =
|
135.74
|
|
corr(u_i, X) = 0 (assumed)
|
|
Prob > chi2 =
|
0.0000
|
|
lprod |
|
Coef.
|
Std. Err.
|
z
|
P>|z|
|
[95% Conf.
|
Interval]
|
|
+
|
|
|
|
|
|
|
|
lsup |
|
.4453965
|
.0720188
|
6.18
|
0.000
|
.3042423
|
.5865507
|
|
lpoprur |
|
.5582101
|
.1141239
|
4.89
|
0.000
|
.3345313
|
.7818889
|
|
|
|
e
|
Réalisé et soutenu par Samson James Aimé
AGBO et Rodrigue Noutaï HONKPEHEDJI
|
|
|
lprix | -.2161175 .1321722 -1.64 0.102 -.4751703
.0429353
lhautpl | -.117592 .1222238 -0.96 0.336 -.3571462
.1219621
_cons | -.4443465 1.535815 -0.29 0.772 -3.454488
2.565795
+
sigma_u | .55216822
sigma_e | .26264837
rho | .81548815 (fraction of variance due to
u_i)
. hausman fixed
---- Coefficients ----
| (b) (B) (b-B) sqrt(diag(V_b-V_B))
| fixed . Difference S.E.
+
lsup | .399503 .4453965 -.0458934 .
lpoprur | .5146713 .5582101 -.0435389
.0084152
lprix | -.2096241 -.2161175 .0064934 .
lhautpl | -.1152291 -.117592 .0023629 .
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(4) =
(b-B)'[(V_b-V_B)^(-1)](b-B)
= 10.78
Prob>chi2 = 0.0292
(V_b-V_B is not positive definite)
predict fixed, u sktest fixed
Skewness/Kurtosis tests for Normality
joint
Variable | Pr(Skewness) Pr(Kurtosis) adj chi2(2)
Prob>chi2
+
fixed | 0.001 0.373 10.51 0.0052
. gen fixed2 = fixed*fixed
. reg fixed2 lprod lsup lpoprur lprix lhautpl, fe option
fe not allowed
r(198);
. reg fixed2 lsup lpoprur lprix lhautpl, fe option fe not
allowed
r(198);
. reg fixed2 lsup lpoprur lprix lhautpl
Source | SS df MS Number of obs = 88
+ F( 4, 83) = 8.53
Model | 10.1582918 4 2.53957294 Prob > F =
0.0000
Residual | 24.7146116 83 .297766405 R-squared =
0.2913
+ Adj R-squared = 0.2571
Total | 34.8729034 87 .40083797 Root MSE =
.54568
fixed2 | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
+
lsup | -.1340434 .082597 -1.62 0.108 -.2983254
.0302387
lpoprur | .5297644 .127636 4.15 0.000 .2759016
.7836273
|
|
|
f
|
Réalisé et soutenu par Samson James Aimé
AGBO et Rodrigue Noutaï HONKPEHEDJI
|
|
|
lprix | -.374556 .2414996 -1.55 0.125 -.854889
.105777
lhautpl | -.6337519 .1865383 -3.40 0.001 -1.004769
-.2627348
_cons | 1.906683 1.969337 0.97 0.336 -2.010248
5.823615
. xtregar lprod lsup lpoprur lprix lhautpl,
fe
FE (within) regression with AR(1) disturbances Number of
obs = 80
Group variable: comm Number of groups = 8
R-sq: within = 0.5400 Obs per group: min = 10
between = 0.9095 avg = 10.0
overall = 0.8667 max = 10
F(4,68) = 19.96
corr(u_i, Xb) = 0.7855 Prob > F = 0.0000
lprod | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
+
lsup | .3024995 .0761919 3.97 0.000 .1504609
.454538
lpoprur | .5682697 .1119369 5.08 0.000 .3449032
.7916363
lprix | -.2498191 .126181 -1.98 0.052 -.5016093
.0019712
lhautpl | -.1678023 .1355912 -1.24 0.220 -.4383704
.1027657
_cons | .8956003 1.036103 0.86 0.390 -1.171911
2.963111
+
rho_ar | .34613866
sigma_u | .84524194
sigma_e | .26181561
rho_fov | .91245333 (fraction of variance because of
u_i)
F test that all u_i=0: F(7,68) = 17.80 Prob > F =
0.0000
. xtregar lprod lsup lpoprur lprix lhautpl, fRe option
fRe not allowed
r(198);
. xtregar lprod lsup lpoprur lprix lhautpl,
re
RE GLS regression with AR(1) disturbances Number of obs =
88
Group variable: comm Number of groups = 8
R-sq: within = 0.5767 Obs per group: min = 11
between = 0.9215 avg = 11.0
overall = 0.8793 max = 11
Wald chi2(5) = 170.99
corr(u_i, Xb) = 0 (assumed) Prob > chi2 =
0.0000
lprod | Coef. Std. Err. z P>|z| [95% Conf.
Interval]
+
lsup | .4855914 .0768424 6.32 0.000 .334983
.6361997
lpoprur | .6240351 .1136918 5.49 0.000 .4012032
.8468669
lprix | -.2956279 .1390389 -2.13 0.033 -.5681392
-.0231167
lhautpl | -.1572632 .1376602 -1.14 0.253 -.4270722
.1125459
_cons | -.7440033 1.554131 -0.48 0.632 -3.790045
2.302038
+
rho_ar | .34613866 (estimated autocorrelation
coefficient)
sigma_u | .35540846 sigma_e | .32537836 rho_fov |
.5440252 (fraction of variance due to u_i)
theta | .62604963
|
|
|
g
|
Réalisé et soutenu par Samson James Aimé
AGBO et Rodrigue Noutaï HONKPEHEDJI
|
|
|
|
. xtreg lprod lsup lpoprur lprix lhautpl, re
|
|
|
|
Random-effects GLS regression
|
Number of obs =
|
88
|
|
Group variable:
|
comm
|
|
Number of groups =
|
8
|
|
R-sq: within =
|
0.5770
|
|
Obs per group: min =
|
11
|
|
between =
|
0.9211
|
|
avg =
|
11.0
|
|
overall =
|
0.8795
|
|
max =
|
11
|
|
Random effects u_i ~ Gaussian
|
Wald chi2(4) =
|
135.74
|
|
corr(u_i, X) = 0 (assumed)
|
Prob > chi2 =
|
0.0000
|
|
lprod |
|
Coef.
|
Std. Err. z
|
P>|z| [95% Conf.
|
Interval]
|
|
+
lsup |
|
.4453965
|
.0720188 6.18
|
0.000 .3042423
|
.5865507
|
|
lpoprur |
|
.5582101
|
.1141239 4.89
|
0.000 .3345313
|
.7818889
|
|
lprix |
|
-.2161175
|
.1321722 -1.64
|
0.102 -.4751703
|
.0429353
|
|
lhautpl |
|
-.117592
|
.1222238 -0.96
|
0.336 -.3571462
|
.1219621
|
|
|
_cons
|
-.4443465
|
1.535815 -0.29
|
0.772 -3.454488
|
2.565795
|
|
+
sigma_u |
|
.55216822
|
|
|
|
|
sigma_e |
|
.26264837
|
|
|
|
|
rho |
|
.81548815
|
(fraction of variance due to u_i)
|
|
|
. xtregar lprod lsup lpoprur lprix lhautpl,
re
|
|
|
|
RE GLS regression with AR(1)
|
disturbances
|
Number of obs =
|
88
|
|
Group variable:
|
comm
|
|
Number of groups =
|
8
|
|
R-sq: within =
|
0.5767
|
|
Obs per group: min =
|
11
|
|
between =
|
0.9215
|
|
avg =
|
11.0
|
|
overall =
|
0.8793
|
|
max =
|
11
|
|
|
|
Wald chi2(5) =
|
170.99
|
|
corr(u_i, Xb)
|
= 0 (assumed)
|
Prob > chi2 =
|
0.0000
|
|
lprod |
|
Coef.
|
Std. Err. z
|
P>|z| [95% Conf.
|
Interval]
|
|
+
lsup |
|
.4855914
|
.0768424 6.32
|
0.000 .334983
|
.6361997
|
|
lpoprur |
|
.6240351
|
.1136918 5.49
|
0.000 .4012032
|
.8468669
|
|
lprix |
|
-.2956279
|
.1390389 -2.13
|
0.033 -.5681392
|
-.0231167
|
|
lhautpl |
|
-.1572632
|
.1376602 -1.14
|
0.253 -.4270722
|
.1125459
|
|
|
_cons
|
-.7440033
|
1.554131 -0.48
|
0.632 -3.790045
|
2.302038
|
|
+
|
|
|
|
|
|
rho_ar |
|
.34613866
|
(estimated autocorrelation coefficient)
|
|
|
sigma_u |
|
.35540846
|
|
|
|
sigma_e |
|
.32537836
|
|
|
|
rho_fov |
|
.5440252
|
(fraction of variance due to u_i)
|
|
|
theta |
|
.62604963
|
|
|
|
|
. xtregar lprod lsup lpoprur lprix lhautpl,
fe
|
lbi
|
|
|
FE (within) regression with AR(1)
|
disturbances
|
Number of obs =
|
80
|
|
Group variable: comm
|
|
Number of groups =
|
8
|
|
R-sq: within = 0.5400
|
|
Obs per group: min =
|
10
|
|
between = 0.9095
|
|
avg =
|
10.0
|
|
overall = 0.8667
|
|
max =
|
10
|
|
|
F(4,68) =
|
19.96
|
|
corr(u_i, Xb) = 0.7855
|
|
Prob > F =
|
0.0000
|
|
|
|
h
|
Réalisé et soutenu par Samson James Aimé
AGBO et Rodrigue Noutaï HONKPEHEDJI
|
|
|
|
lprod
|
|
|
Coef.
|
Std. Err.
|
t
|
P>|t|
|
[95% Conf.
|
Interval]
|
|
+
|
|
|
|
|
|
|
|
|
lsup
|
|
|
.3024995
|
.0761919
|
3.97
|
0.000
|
.1504609
|
.454538
|
|
lpoprur
|
|
|
.5682697
|
.1119369
|
5.08
|
0.000
|
.3449032
|
.7916363
|
|
lprix
|
|
|
-.2498191
|
.126181
|
-1.98
|
0.052
|
-.5016093
|
.0019712
|
|
lhautpl
|
|
|
-.1678023
|
.1355912
|
-1.24
|
0.220
|
-.4383704
|
.1027657
|
|
_cons
|
|
|
.8956003
|
1.036103
|
0.86
|
0.390
|
-1.171911
|
2.963111
|
|
+
|
|
|
|
|
|
|
|
|
rho_ar
|
|
|
.34613866
|
|
|
|
|
|
|
sigma_u
|
|
|
.84524194
|
|
|
|
|
|
|
sigma_e
|
|
|
.26181561
|
|
|
|
|
|
|
rho_fov
|
|
|
.91245333
|
(fraction of variance because
|
of u_i)
|
|
F test that all u_i=0: F(7,68) = 17.80 Prob > F =
0.0000
modified Bhargava et al. Durbin-Watson =
1.3889366
Baltagi-Wu LBI = 1.6300891
| 
