Sergio I Prada <[email protected]>:
Did you mean sd or SE in parentheses? I will assume you want the
standard deviations, since that is harder. A simple method requires
editing the tables you get into one table. A more complex method
would be to create a -program- that posts the info where -esttab- or
an alternative looks for it. A better way might be to use -file- to
write out the info in exactly the way you need to save it. Here's the
simple method I had in mind:
ssc inst estout
clear all
webuse nhanes2f
svyset psuid [pw=finalwgt], strata(stratid)
g male=1-female
la var male "Male"
la var female "Female"
svy: reg height male female, nocons
est sto Height
su height [aw=finalwgt] if female==0
mat sd=r(sd)
su height [aw=finalwgt] if female==1
mat sd=sd,r(sd)
mat rownames sd=y1
mat colnames sd=male female
estadd matrix sd=sd
svy: reg weight male female, nocons
est sto Weight
su weight [aw=finalwgt] if female==0
mat sd=r(sd)
su weight [aw=finalwgt] if female==1
mat sd=sd,r(sd)
mat rownames sd=y1
mat colnames sd=male female
estadd matrix sd=sd
esttab Height Weight using twoparts.doc, mti nogaps rtf nostar aux(sd) la
la var female "Difference"
svy: reg height female
est sto Height
svy: reg weight female
est sto Weight
esttab Height Weight using twoparts.doc, mti nogaps rtf la a
On Tue, Jan 26, 2010 at 11:54 AM, Sergio I Prada <[email protected]> wrote:
> Hi, I am looking for help in automating a rather tedious task.
>
> I want to have a table comparing the means for some variables for two
> subpopulations in my survey data, including a test on the diff btw the
> means.
>
> For instance, according to the SVY Manual the right way to get the numbers
> is:
>
> use http://www.stata-press.com/data/r11/nhanes2f
> svyset psuid [pweight=finalwgt], strata(stratid)
> svy: mean height, over(sex)
> lincom [height]Male - [height]Female
> svy: mean weight, over(sex)
> lincom [weight]Male - [weight]Female
>
>
> The table I'd like to have looks like
>
> Male Female Diff
> mena/sd mean/sd
>
> Height 175.58 161.90 13.67***
> (0.17) (0.13)
> Weight 78.63 65.71 12.92***
> (0.21) (0.27)
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
