Annexe 5 : Test de Hausman [équation
(12)]
---- Coefficients ----
| (b) (B) (b-B)
sqrt(diag(V_b-V_B))
| fixe . Difference
S.E.
-------------+----------------------------------------------------------------
assets | .3605072 .0465768 .3139305
.1421929
capital | .3342695 -.0666735 .400943
.3419486
costs | .3540308 .3087626 .0452682
.2660831
deposits | -.0349325 -.034633 -.0002996
.0448837
loans | -.6839437 -.7860789 .1021351
.2603835
number_of_~s | -.0436093 -.0340541 -.0095553
.0504573
provisions | .0240529 .1143432 -.0902904
.1130979
reserves | .2398257 .1831751 .0566506
.0802878
dummy_2002 | 2.585148 2.084322 .5008259
1.704532
dummy_2003 | 2.401175 1.78951 .6116655
1.806853
dummy_2004 | .5772514 .1946533 .3825981
1.433704
dummy_2005 | .9216012 -.1545392 1.07614
1.173777
dummy_2006 | 2.110732 1.013866 1.096866
.8504032
dummy_2007 | .2659551 .1446733 .1212818
.3565814
------------------------------------------------------------------------------
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(14) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 6.69
Prob>chi2 = 0.9461
(V_b-V_B is not positive definite)
Source : Construit par l'auteur.
Annexe 6 : Test de significativité des
effets aléatoires [équation (12)]
Breusch and Pagan Lagrangian multiplier test for random
effects:
spread[country,t] = Xb + u[country] + e[country,t]
Estimated results:
| Var sd = sqrt(Var)
---------+-----------------------------
spread | 3.749768 1.936432
e | 1.919333 1.3854
u | 0 0
Test: Var(u) = 0
chi2(1) = 2.10
Prob > chi2 = 0.1471
Source : Construit par l'auteur.
Annexe 7 : Estimation du modèle à
effets fixes [équation (13)]
Fixed-effects (within) regression Number of obs
= 42
Group variable (i): country Number of
groups = 6
R-sq: within = 0.1644 Obs per group:
min = 7
between = 0.4003
avg = 7.0
overall = 0.0306
max = 7
F(3,33)
= 2.16
corr(u_i, Xb) = -0.3394 Prob > F
= 0.1109
------------------------------------------------------------------------------
pure_spread | Coef. Std. Err. t P>|t|
[95% Conf. Interval]
-------------+----------------------------------------------------------------
gdp_growth | -.0286054 .0215318 -1.33 0.193
-.0724122 .015201
inflation_~e | .068334 .0408616 1.67 0.104
-.0147995 .1514675
ir_volatil~y | -.0278175 .0380687 -0.73 0.470
-.1052689 .0496338
_cons | 10.55085 .3329708 31.69 0.000
9.873412 11.22828
-------------+----------------------------------------------------------------
sigma_u | 2.4244433
sigma_e | .7950325
rho | .90290684 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(5, 33) = 43.97 Prob
> F = 0.0000
Source : Construit par l'auteur.
Prob > F = 0.0000 0.05 on rejette H0 et on conclu
que les effets individuels sont nécessaires. Il convient à
présent de déterminer quel est le type d'effets individuels
à prendre en compte.
| 
