Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: data management


From   "R.E. De Hoyos" <redeho2@hotmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: data management
Date   Wed, 18 Feb 2004 12:10:35 -0000

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" <Richard.A.Williams.5@nd.edu>
To: <statalist@hsphsun2.harvard.edu>
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:  Richard.A.Williams.5@ND.Edu
> 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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index