Row names with esttab using over()

[Alex Taylor]

Hi all,

I'm using Stata 17 and the user-written esttab command from SSC to create/export tables to rtf. I need to use the mean command with the over suffix because I'm working with complex survey data. My issue is getting the correct format for the names of the rows. Here's an illustration using the auto data.

sysuse auto, clear
eststo: mean weight, over(foreign)

This prints a table in the display with the row names determined by the labels for foreign, i.e. "Domestic" and "Foreign". However, when I utilize the esttab command,

esttab, wide se nostar

the row names are "[email protected]" and "[email protected]". The question is how do I get the row names as "Domestic" and "Foreign"?

I recognize this can be done manually like so:

esttab, wide se nostar varlabels([email protected] "Domestic" [email protected] "Foreign")

However, I'm dealing with a very large number of variables so need an automated solution. Thanks in advance.

[Andrew Musau]

esttab takes whatever is in e(b), and mean does not store the coefficients as you expect - so there is not much that you can do about that. I would suggest that you use regress in place of mean.

webuse nhanes2f
svyset psuid [pweight=finalwgt], strata(stratid)
svy: mean zinc, over(race)
svy: regress zinc ibn.race, nocons
esttab, wide se nostar label

Res.:

. svy: mean zinc, over(race)
(running mean on estimation sample)

Survey: Mean estimation

Number of strata =      31      Number of obs   =        9,189
Number of PSUs   =      62      Population size =  104,176,071
                                Design df       =           31

--------------------------------------------------------------
             |             Linearized
             |       Mean   Std. Err.     [95% Conf. Interval]
-------------+------------------------------------------------
 c.zinc@race |
      White  |   87.49539   .4791963      86.51806    88.47272
      Black  |   85.08574   1.165209      82.70929     87.4622
      Other  |   83.57091   1.585463      80.33734    86.80448
--------------------------------------------------------------

.
. svy: regress zinc ibn.race, nocons
(running regress on estimation sample)

Survey: Linear regression

Number of strata   =        31                Number of obs     =        9,189
Number of PSUs     =        62                Population size   =  104,176,071
                                              Design df         =           31
                                              F(   3,     29)   =     10616.73
                                              Prob > F          =       0.0000
                                              R-squared         =       0.9723

------------------------------------------------------------------------------
             |             Linearized
        zinc |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        race |
      White  |   87.49539   .4791963   182.59   0.000     86.51806    88.47272
      Black  |   85.08574   1.165209    73.02   0.000     82.70929     87.4622
      Other  |   83.57091   1.585463    52.71   0.000     80.33734    86.80448
------------------------------------------------------------------------------

.
. esttab, wide se nostar label

----------------------------------------------
                              (1)            
                     serum zinc~)            
----------------------------------------------
White                       87.50      (0.479)
Black                       85.09      (1.165)
Other                       83.57      (1.585)
----------------------------------------------
Observations                 9189            
----------------------------------------------
Standard errors in parentheses

.

[Alex Taylor]

Thanks, Andrew. Great idea. That should work!