Stacking tabulate tables into a large table with est

[Chris Martin]

I've successfully generate a set of descriptive tables for six variables, and I can manually stack them on top of each other, but would like to know if there's a way to automate this. This is what my code currently looks like:

. eststo clear
. eststo: quietly estpost svy: tabulate ageG22 if Md1sample==1
. esttab, cell("b(f(4))") varlabels(`e(labels)') varwidth(16)
. eststo: quietly estpost svy: tabulate genderC if Md1sample==1
. esttab, cell("b(f(4))") varlabels(`e(labels)') varwidth(16)
. quietly estpost svy: tabulate raceD if Md1sample==1
. esttab, cell("b(f(4))") varlabels(`e(labels)') varwidth(16)
. quietly estpost svy: tabulate relcat if Md1sample==1
. esttab, cell("b(f(4))") varlabels(`e(labels)') varwidth(16)
. quietly estpost svy: tabulate educ_par1 if Md1sample==1
. esttab, cell("b(f(4))") varlabels(`e(labels)') varwidth(25)
. quietly estpost svy: tabulate educ_par1 if Md1sample==1
. esttab, cell("b(f(4))") varlabels(`e(labels)') varwidth(25)

ageG22 genderC raceD and so on are the variables being tabulated. Ideally, the column total would be eliminated from each tabulation, and I would get one long table with value label on the left and the proportion on the right.

Thanks
Chris

[Oded Mcdossi]

Here is an example of what you are looking for:

webuse nhanes2f
svyset psuid [pweight=finalwgt], strata(stratid)
eststo clear
foreach i of var agegrp female race {
 eststo: estpost svy: tabulate `i' 
 esttab using myfile.csv, cell("b(f(4))") varlabels(`e(labels)') varwidth(16) app
 eststo clear 
}

[Chris Martin]

Thank you. Is there a way to suppress (1), "Mean" "b", and the total on the right side; and add the variable name on the left side? Currently the first few rows of the Excel sheet look like this

(1)
Mean
b
Male 0.4424
Female 0.5480
Trans/Other 0.0096
Total 1.0000
N 18527
(1)
Mean
b
Non-Hisp. White 0.6825
Hisp. White 0.0520
Hispanic Only 0.0293

I'm not sure if it's possible but I'd like to generate something like this:

Gender
Male 0.4424
Female 0.5480
Trans/Other 0.0096
Race
Non-Hisp. White 0.6825
Hisp. White 0.0520
Hispanic Only 0.0293

[Chris Martin]

Sorry I realize that didn't post correctly. Here's what I meant to paste:

                    (1)
                    Mean
                    b
Male                0.4424
Female              0.5480
Trans/Other         0.0096
Total               1.0000
N                   18527
                    (1)
                    Mean
                    b
Non-Hisp. White     0.6825
Hisp. White         0.0520
Hispanic Only       0.0293

And this is what I want:

Gender    
Male                0.4424
Female              0.5480
Trans/Other         0.0096
Race    
Non-Hisp. White     0.6825
Hisp. White         0.0520
Hispanic Only       0.0293

[Oded Mcdossi]

I think that will do the trick. You should read the help file of esttab and this website where you'll find many examples:

webuse nhanes2f, clear
svyset psuid [pweight=finalwgt], strata(stratid)

label var agegrp "Age groups"
label var sex "Sex"
label var race "Race"
eststo clear

foreach i of var agegrp sex race {
 eststo: estpost svy: tabulate `i' 
 esttab using myfile.csv, cells("b(label(`: var label `i'')fmt(4))")  mtitle("") nonumber noobs app
 eststo clear 
}

[Chris Martin]

Thanks. That takes away the serial number and "Mean" but still leaves b and there's no column title. I also get the recurring message

row labels saved in macro e(labels)
(est1 stored)
nothing found where name expected
(output written to myfile.csv)

Despite the message I do get a csv file that's easier to manually edit. I've read the esttab website but I'll have a look again.

[Oded Mcdossi]

I can't reproduce your example. When I run the code in #5 this is the results I get, which is very similar for what you are asking for:

 

Age groups

age20-29
0.2808

age30-39
0.2045

age40-49
0.1683

age50-59
0.1670

age60-69
0.1332

age 70+
0.0462

Total
1.0000





Sex

Male
0.4796

Female
0.5204

Total
1.0000





Race

White
0.8790

Black
0.0956

Other
0.0254

Total
1.0000

Please show us exactly what you typed and what Stata gave you back.