Tabout: produce a table of means with multiple variables

[Giorgio Gerini Romagnoli]

Dear Statalisters,

This is my first post, therefore I hope it will be comprehensible and compliant to the rules.

I would like to use the command tabout in Stata16 on Windows10 to generate a table containing the mean of 5 variables.
In order to be more clear, I'd like to make an example with a common dataset.

> sysuse auto
> tabulate foreign rep78, summarize (price) means
> tabout foreign rep78 using giorgio.txt, replace c(mean price) sum format (2 2) clab(_ _) layout(rb) h3(nil) style(tex) bt font(bold) topstr(11cm)

What I have here is quite clear and simple for me, I cross-tabulated foreign and rep78 with the table containing values of the mean of price.

What if I would be interested in tabulating and then exporting a table which contains not only the mean of price but, for example, also the mean of the weight of the cars?

As I understand I am not good with words, I am gonna show below a short example of what I would need.

	repair record 1978	1	2	3	4	5		total
	car type
mean of price	domestic
	foreign
mean of weight	domestic
	foreign

Do you think this is possible?

Thank you in advance for all your answers and I hope I was clear enough

Giorgio

[Andrew Musau]

I do not use tabout (SSC), but here is a way to get what you want in a dataset. You can export in whichever format you want.

*WANTED
sysuse auto, clear
foreach var in price weight mpg{
    tabulate foreign rep78, summarize (`var') means
}

*CODE:
clear
tempfile a1
gen var=""
save `a1'
sysuse auto, clear
foreach var in mpg weight price{
    separate `var', by(rep78) gen(cat)
    drop if missing(rep78)
    egen total= rowtotal(cat?)
    collapse cat? total, by(foreign)
    gen var= "`var'"
    append using `a1'
    save `a1', replace
    sysuse auto, clear
}
use `a1', clear
order var
l, sepby(var)

Res.:

. foreach var in price weight mpg{
  2.
.     tabulate foreign rep78, summarize (`var') means
  3.
. }


                              Means of Price

           |                Repair Record 1978
  Car type |         1          2          3          4          5 |     Total
-----------+-------------------------------------------------------+----------
  Domestic |   4,564.5  5,967.625  6,607.074  5,881.556    4,204.5 |  6,179.25
   Foreign |         .          .  4,828.667  6,261.444  6,292.667 | 6,070.143
-----------+-------------------------------------------------------+----------
     Total |   4,564.5  5,967.625  6,429.233    6,071.5      5,913 | 6,146.043

                          Means of Weight (lbs.)

           |                Repair Record 1978
  Car type |         1          2          3          4          5 |     Total
-----------+-------------------------------------------------------+----------
  Domestic |     3,100   3,353.75  3,442.222  3,532.222      1,960 | 3,368.333
   Foreign |         .          .      2,010  2,207.778  2,403.333 | 2,263.333
-----------+-------------------------------------------------------+----------
     Total |     3,100   3,353.75      3,299      2,870  2,322.727 | 3,032.029

                          Means of Mileage (mpg)

           |                Repair Record 1978
  Car type |         1          2          3          4          5 |     Total
-----------+-------------------------------------------------------+----------
  Domestic |        21     19.125         19  18.444444         32 | 19.541667
   Foreign |         .          .  23.333333  24.888889  26.333333 | 25.285714
-----------+-------------------------------------------------------+----------
     Total |        21     19.125  19.433333  21.666667  27.363636 | 21.289855


. l, sepby(var)

     +--------------------------------------------------------------------------------+
     |    var    foreign      cat1      cat2      cat3      cat4      cat5      total |
     |--------------------------------------------------------------------------------|
  1. |  price   Domestic   4,564.5   5,967.6   6,607.1   5,881.6   4,204.5    6179.25 |
  2. |  price    Foreign         .         .   4,828.7   6,261.4   6,292.7   6070.143 |
     |--------------------------------------------------------------------------------|
  3. | weight   Domestic     3,100   3,353.8   3,442.2   3,532.2     1,960   3368.333 |
  4. | weight    Foreign         .         .     2,010   2,207.8   2,403.3   2263.333 |
     |--------------------------------------------------------------------------------|
  5. |    mpg   Domestic        21    19.125        19   18.4444        32   19.54167 |
  6. |    mpg    Foreign         .         .   23.3333   24.8889   26.3333   25.28572 |
     +--------------------------------------------------------------------------------+

[Giorgio Gerini Romagnoli]

Thank you very much for your answer Andrew, I would never have been able to produce a code like that.

[Giorgio Gerini Romagnoli]

Dear Andrew, I tried in these days to export the table on PNG. If I have to be honest, I am not able to work with the temporary file.

[Andrew Musau]

You can export as a MS Word or MS Excel file. Here is a way to do the former using putdocx. The code is modified to use variable labels and some variables are renamed.

tempfile a1
gen Variable=""
save `a1'
sysuse auto, clear
foreach var in mpg weight price{
    gen Variable= "`:var lab `var''"
    separate `var', by(rep78) gen(_)
    drop if missing(rep78)
    egen Total= rowtotal(_?)
    collapse _? Total, by(foreign Var)
    append using `a1'
    save `a1', replace
    sysuse auto, clear
}
use `a1', clear
rename foreign Origin
order Var
putdocx begin
putdocx table mytable = data(*), varnames width(100%)
putdocx save myfile.docx, replace

Res.: