[Sophie Starck]

Do you mean something like this? This is what I get back from Stata when I type in the previously mentioned code.

. do "/var/folders/4n/7c0b28650cq9k5_h6wlc10rr0000gn/T//SD66974.000000"

. qui xtreg g_gdp length g_prim1 g_death patent1 polity2 g_gov_con g_trade g_pop initial_gdp if (country == "Benin" | country == "Burkina Faso" | co
> untry == "Congo" | country == "Algeria" | country == "Gabon" | country == "Kenya" | country == "Algeria" | country == "Gabon" | country == "Kenya"
>  | country == "Madagascar" | country == "Niger" | country == "Togo") & inrange(length, 1, 42), vce(robust)

. qui outreg using Table3.doc, nocons addrow(Country-fixed effects,NO \ Year-fixed effects,NO) se bdec (3) starlevels(10 5 1) sigsymbols(*,**,***) s
> tarloc(1) summstat(r2 \ N) summtitle(R-squared \ Observations) summdec(2 2 0) title(Table 3. \ The Relationship between Social Infrastructure and 
> Economic Growth \ Dependent Variable: GDP Growth) ctitle("", "" \ "Explanatory Variables" \ "" "(1)") blankrows nolegend varlabels replace 

.

[Carlo Lazzaro]

Sophie:
not quite.
I meant the table that Stata gives you back after the code:

 
 xtreg g_gdp length g_prim1 g_death patent1 polity2 g_gov_con g_trade g_pop initial_gdp if (country == "Benin" | country == "Burkina Faso" | co > untry == "Congo" | country == "Algeria" | country == "Gabon" | country == "Kenya" | country == "Algeria" | country == "Gabon" | country == "Kenya" >  | country == "Madagascar" | country == "Niger" | country == "Togo") & inrange(length, 1, 42), vce(robust)

[Sophie Starck]

Okay. This what I get:



Random-effects GLS regression Number of obs =
> 369
Group variable: id Number of groups =
> 9

R-squared: Obs per group:
Within = 0.0927 min =
> 41
Between = 0.0025 avg =
> 41.0
Overall = 0.0902 max =
> 41

Wald chi2(8) =
> .
corr(u_i, X) = 0 (assumed) Prob > chi2 =
> .

(Std. err. adjusted for 9 clust
> ers in id)
--------------------------------------------------------------------
> ----------
| Robust
g_gdp | Coefficient std. err. z P>|z| [95% conf.
> interval]
-------------+------------------------------------------------------
> ----------
length | -.0005948 .0003391 -1.75 0.079 -.0012594
> .0000697
g_prim1 | .1296504 .0351749 3.69 0.000 .060709
> .1985919
g_death | -.1957951 .2610774 -0.75 0.453 -.7074974
> .3159072
patent1 | -.0023113 .0166349 -0.14 0.889 -.0349152
> .0302926
polity2 | -.0003284 .0009512 -0.35 0.730 -.0021928
> .001536
g_gov_con | -.1475374 .0318031 -4.64 0.000 -.2098703
> -.0852046
g_trade | .0519977 .0367851 1.41 0.157 -.0200999
> .1240952
g_pop | .0454166 .7137295 0.06 0.949 -1.353467
> 1.444301
initial_gdp | 1.55e-07 2.81e-07 0.55 0.582 -3.97e-07
> 7.07e-07
_cons | .0356128 .0369794 0.96 0.336 -.0368654
> .108091
-------------+------------------------------------------------------
> ----------
sigma_u | 0
sigma_e | .0879722
rho | 0 (fraction of variance due to u_i)
--------------------------------------------------------------------
> ----------

.

[Carlo Lazzaro]

Sophie:
1) as per your results, you do not have any evidence of a panel-wise effect (sigma_u=0). Pooled OLS is the way to go here.
2) using clustered robust standard errors with only 9 panels does not make sense; switch to their default counterparts instead.

[Sophie Starck]

Okay thank you.
Does the R2 not show when using outreg because the value is too low?

[Carlo Lazzaro]

Sophie:
I'm not familiar with the community-contributed command -outreg-.
That said, there's an alternative option for what you are after: see -table- suite examples in Stata 17 .pdf manual.