I first run the regression for the period 2001 to 2018. Guided by the Hausman test, I have used country fixed effects. Also, standard errors are clustered at the country-level. I find that a 1% increase in growth of EM trade partners is associated with a 0.5% increase in the growth of the reporting country.
Code:
xtreg realgthpc l.tpg_aenia l.tpg_em l.tpg_chn realpcgdp_5yago_bn gfcf_nomgdp prim_enrol cpi_ep_inflatn depratio open_gdp globalgthpc oiltrdbal_gdp oiltrdbal_gdp_5y gth_commexp_prind crisis_dummy moderation_dummy recovery_dummy, fe vce(cluster CountryNo)
Code:
Fixed-effects (within) regression Number of obs = 343
Group variable: CountryNo Number of groups = 26
R-sq: Obs per group:
within = 0.3832 min = 4
between = 0.0080 avg = 13.2
overall = 0.1749 max = 17
F(16,25) = 15.10
corr(u_i, Xb) = -0.5907 Prob > F = 0.0000
(Std. Err. adjusted for 26 clusters in CountryNo)
------------------------------------------------------------------------------------
| Robust
realgthpc | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------------+----------------------------------------------------------------
tpg_aenia |
L1. | -.0420772 .1956026 -0.22 0.831 -.4449283 .3607739
|
tpg_em |
L1. | .5208979 .2481788 2.10 0.046 .009764 1.032032
|
tpg_chn |
L1. | .1076752 .5621433 0.19 0.850 -1.050081 1.265431
|
realpcgdp_5yago_bn | .0309591 .0211014 1.47 0.155 -.0125001 .0744183
gfcf_nomgdp | .2013728 .0805251 2.50 0.019 .0355283 .3672173
prim_enrol | .0969751 .0289507 3.35 0.003 .0373501 .1566001
cpi_ep_inflatn | -.1490856 .0699263 -2.13 0.043 -.2931015 -.0050697
depratio | -.0002485 .0571743 -0.00 0.997 -.1180011 .1175041
open_gdp | .0272978 .023562 1.16 0.258 -.0212292 .0758247
globalgthpc | .7846087 .2894573 2.71 0.012 .1884603 1.380757
oiltrdbal_gdp | .061986 .079659 0.78 0.444 -.1020748 .2260467
oiltrdbal_gdp_5y | .0130573 .1709793 0.08 0.940 -.339081 .3651957
gth_commexp_prind | .0252099 .0140975 1.79 0.086 -.0038245 .0542443
crisis_dummy | .134008 .9356604 0.14 0.887 -1.793021 2.061037
moderation_dummy | .8694931 .6094907 1.43 0.166 -.3857765 2.124763
recovery_dummy | .5936562 .6663803 0.89 0.381 -.7787797 1.966092
_cons | -14.92877 6.315646 -2.36 0.026 -27.93609 -1.921454
-------------------+----------------------------------------------------------------
sigma_u | 2.7801935
sigma_e | 2.6945358
rho | .51564218 (fraction of variance due to u_i)
------------------------------------------------------------------------------------
0 Response to Problem with estimation of coefficient in panel regression
Post a Comment