How to place significant stars infront of standard errors but still maintain the wide format

[Ethan Gaala]

Hi,
I'm trying to get an esttab table that looks like table 1 attached. specifically, i want the stars placed after the standard errors which are also placed after the coefficients, all in one horizontal line. However, whenever i run the code and specify wide format, it breaks the lines into two. each. as seen in table 2.
Here's my code, ran on state 16:

logit Approval_of_Protest GPDpc_PPP_log i.Religiousity_2
est store ModelA
xtmelogit Approval_of_Protest GPDpc_PPP_log i.Religiousity_2 || country:, cov(unstr)
est store ModelB

esttab ModelA ModelB using results.rtf, wide cells(b(fmt(3)) se(par fmt(2)star)) replace

* Example generated by -dataex-. To install: ssc install dataex
clear
input byte Approval_of_Protest float GPDpc_PPP_log byte Religiousity_2 str48 country
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  0 "Algeria"
 0 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
.b 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 1 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  1 "Algeria"
 1 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  0 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  0 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
.b 32.622215  1 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215 .a "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  1 "Algeria"
 1 32.622215  1 "Algeria"
 0 32.622215  1 "Algeria"
 1 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  1 "Algeria"
 1 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
.b 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  1 "Algeria"
 0 32.622215  1 "Algeria"
 0 32.622215  1 "Algeria"
 1 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  1 "Algeria"
 1 32.622215  2 "Algeria"
 1 32.622215  1 "Algeria"
 0 32.622215  1 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  0 "Algeria"
 1 32.622215  1 "Algeria"
.b 32.622215 .b "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 1 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
.b 32.622215 .b "Algeria"
 0 32.622215  1 "Algeria"
 0 32.622215  1 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  0 "Algeria"
 0 32.622215  1 "Algeria"
.b 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
 0 32.622215  1 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 0 32.622215  2 "Algeria"
 1 32.622215  2 "Algeria"
end
label values Approval_of_Protest Approval_of_Protest
label def Approval_of_Protest 0 "No", modify
label def Approval_of_Protest 1 "Yes", modify
label values Religiousity_2 Religiousity_2
label def Religiousity_2 0 "Not Religious", modify
label def Religiousity_2 1 "Somewhat Religious", modify
label def Religiousity_2 2 "Highly Religious", modify

------------------ copy up to and including the previous line ------------------

[Andrew Musau]

estout is from Stata Journal, as you are asked to explain (FAQ Advice #12). The option that you are missing is -staraux- rather than defining two cells separately. Secondly, rtf format will not allow you to have a wide table, so you need to use format "doc" for MS Word. In general, if you want a multi-column table spanning 3+ columns, you need to create a latex table as long wide tables exported to MS Word will be split into multiple lines. Up to 3 columns + variable names is fine.

sysuse auto, clear
local vars headroom trunk weight foreign
eststo clear
forval i=1/ `=wordcount("`vars'") '{
    eststo: regress mpg `=word("`vars'", 1)' - `=word("`vars'", `i')'
}
esttab est1 est2 est3 using myfile.doc, aux(se 2) main(b 3) staraux ///
noobs nonumber mlabels("Model 1" "Model 2" "Model 3") ///
collabels(none) replace onecell alignment(l) gaps

Res.:

------------------------------------------------------------
                  Model 1         Model 2         Model 3  
------------------------------------------------------------
headroom     -2.830 (0.73)*** -0.350 (0.88)    0.005 (0.64)  
 
trunk                        -0.741 (0.17)*** -0.097 (0.15)  
 
weight                                       -0.006 (0.00)***
 
_cons        29.768 (2.28)*** 32.537 (2.15)*** 39.684 (1.81)***
------------------------------------------------------------
b coefficients; se in parentheses
* p<0.05, ** p<0.01, *** p<0.001

[Ethan Gaala]

Thank you Andrew Musau for this. I just learned something new and it's saved me a lot of time.