Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Categorical variable models - weight option


From   Steve Samuels <sjsamuels@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Categorical variable models - weight option
Date   Tue, 27 Jul 2010 16:03:42 -0400

If weights are provided with the database (references, please, as the
FAQ request, as many people, like me, will not know what it is)., they
will ordinarily be -pweights- and you should -svyset- your data to
account for the design. .  Consult the study documentation for the
design.

; I doubt if it states that the weights are "analytic". Often, but not
always, -aweights- and -iweights- will give identical results.
  You must use the weights, or your estimates will be biased, compared
to what you would get if you have a simplerandom sample.  If there was
multistage sampling, then your standard errors are incorrect if you do
not specify the design with -svyset-.

Steve

On Tue, Jul 27, 2010 at 3:20 PM, Leonor Saravia <lmisaravia@gmail.com> wrote:
> Dear Statalisters,
>
> Currently, I am working with social protection database, with which I
> have estimate different models, like OLS and 2SLS. Ussualy, when I run
> these kind of models I use the analytic weights ('aweight') to expand
> the sample to the population estimates. Now, I am performing
> categorical variable models: logit, probit, multinomial logit and
> multinomial probit, but when I try to use the 'aweight' option on the
> probit or logit estimation, Stata brings this error: "aweight not
> allow". However, when I run the mlogit or mprobit (usig Stata 10), the
> regression runs well.
>
> I think that the option in this kind of models is to use the 'pweight'
> option, but I do not know if this is correct or if it is a better
> option not to use weights . I would greatly appreciate your
> suggestions and advice on applying the "weight" option.
>
> Kind regards,
>
> Leonor Saravia
> *
> *   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/
>



-- 
Steven Samuels
sjsamuels@gmail.com
18 Cantine's Island
Saugerties NY 12477
USA
Voice: 845-246-0774
Fax:    206-202-4783

*
*   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/


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