You are right Richard, you cannot set analytic weights using the survey
command.
However, as you pointed out, the -svyset- in particular pweights will
provide the correct /point/ estimates (same as using aweights); the
difference (aw vs. pw) being standard errors.
rafa
----- Original Message -----
From: "Richard Williams" <[email protected]>
To: <[email protected]>
Sent: Wednesday, February 18, 2004 11:49 AM
Subject: Re: st: data management
> At 10:15 AM 2/18/2004 +0000, R.E. De Hoyos wrote:
> >Is there any way I can use weights without including
> >it in the tabstat command? I want to run a bunch of
> >summary table, would like to use weights for all of
> >them.
> >
> >Try the svy commands (help -svy-). You can set different weights
> >using -svyset- and once you've done this Stata will use them as long as
you
> >use the prefix -svy- with other valid commands. For example you can
> >use -svytab- or -svysum- and it will give you two-way tables and summary
> >statistics of survey data (or other type needing re-weighting)
respectively.
>
> Good idea for many cases -- but in this case the desire was to use
> aweights, which are not included in the svy commands, are they?
>
> To educate myself here -- In practice, am I correct in saying that using
> the svy commands would force the use of robust standard errors, whereas
you
> would not need to use robust standard errors if you used the aw parameter
> on each separate command?
>
> Put another way, am I correct in saying that
>
> [aw = wgt], robust
>
> and
>
> [pw = wgt]
>
> and
>
> [pw = wgt], robust
>
> all produce the same results? i.e. aw and pw basically do the same
> adjustments, but pw forces the use of robust standard errors?
>
> Of course, for something like tables, maybe it doesn't really matter
> whether you use aw or pw, i.e. if he uses the svytab command using pw=wgt,
> he'll get the same results as if he had used tabstat with aw = wgt?
>
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> FAX: (574)288-4373
> HOME: (574)289-5227
> EMAIL: [email protected]
> WWW (personal): http://www.nd.edu/~rwilliam
> WWW (department): http://www.nd.edu/~soc
>
> *
> * 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/
>
*
* 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/
