Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Weighted Tables


From   "Austin Nichols" <[email protected]>
To   [email protected]
Subject   Re: st: Weighted Tables
Date   Fri, 24 Feb 2006 10:28:03 -0500

I don't see that anyone else has replied, so
. tab prtvtch lnghoma [aw=dweight], row col cell nofr
is what you want, I think.  But you could find out about all the
relevant options easily enough by typing -help tab- and -help svy- so
perhaps that is why no one has replied.

Note that frequencies are not what you want--hence the -nofr-
option--these will not be the same with [aweight] and [pweight] and
will in general probably not measure what you think of as frequencies.
 For an illustration, try:
. tab prtvtch lnghoma
. tab prtvtch lnghoma [aw=dweight]
. svyset [pw=dweight]
. svy: tab prtvtch lnghom, count

On 2/23/06, David Lublin <[email protected]> wrote:
> Does this just mean that I could also just do
>
> . tab prtvtch lnghoma [aw=dweight]
>
> and I would get the same results as if pweight were allowed with tables?
>
> Thanks also for your suggestions of add-ons--saves me the trouble of
> calculating percentages!
>
> David
>
>
> "Austin Nichols" <[email protected]>
> After reading -help svy-, also try
>
>  . tab prtvtch lnghoma [aw=dweight], row col nofr
>
> since the point estimates are identical using aweights or pweights
> (only the standard errors change, as [pw]=[aw]+_robust, roughly), and
> the -svy- commands can be much slower in big datasets.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index