Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Categorical variable models - weight option
From
Steve Samuels <[email protected]>
To
[email protected]
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 <[email protected]> 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
[email protected]
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/