Linebreaks in the mlabes Option for esttab.

[Johannes Hart]

Hello,

I have been looking for an answer in this forum for a long time now, but have not found a working solution.

my question is actually very simple. I want to create a regression table with the esttab command and modify the titles of the columns using mlabes. However, the titles are relatively long and so I need to insert a line break to make them visible.

How can I do this?

[Andrew Musau]

estout is from SSC (FAQ Advice #12). If you are creating LaTeX tables, see https://www.statalist.org/forums/for...tles-in-esttab. Also, if you have a maximum of 2 lines, see other suggestions further down that thread. The following solution is unconventional but may be able to give you what you want.

sysuse auto, clear
regress mpg weight turn disp i.rep78
esttab ., mlabel("`=char(13)'`=char(10)' A very very long label for outcome `=char(13)'`=char(10)' mileage and regressors weight, turn, `=char(13)'`=char(10)' displacement and repair record") noabbrev nobaselevels

Res.:

. esttab ., mlabel("`=char(13)'`=char(10)' A very very long label for outcome `=char(13)'`=char(10)' mileage an
> d regressors weight, turn, `=char(13)'`=char(10)' displacement and repair record") noabbrev nobaselevels

----------------------------
                      (1)  
            
 A very very long label for outcome
 mileage and regressors weight, turn,
 displacement and repair record  
----------------------------
weight           -0.00543**
                  (-3.13)  

turn               -0.187  
                  (-0.92)  

displacement      0.00720  
                   (0.56)  

2.rep78            -0.421  
                  (-0.15)  

3.rep78            -0.754  
                  (-0.29)  

4.rep78            -0.963  
                  (-0.36)  

5.rep78             1.714  
                   (0.62)  

_cons               44.13***
                   (6.67)  
----------------------------
N                      69  
----------------------------
t statistics in parentheses
* p<0.05, ** p<0.01, *** p<0.001