Ignoring the strata might lead to larger standard errors, but will
otherwise not affect the validity of your inference. If some of the
strata are based on design factors like "urban/rural" or "region" you
might add these to your model.
-Steve
On Mon, Jan 25, 2010 at 4:44 PM, Ana <[email protected]> wrote:
> Dear Statalisters,
>
> My study is based on a complex sample design survey and I use a
> switching regression model.
> If I use the 'movestay' command with pweight, my estimates will be
> unbiased and consistent.
> However, the standard errors will be based on the weights and clusters
> and it will not take account the stratification of the survey.
> Is statistically acceptable to run the 'movestay' command without
> taking account the stratification of the survey?
>
> Best regards,
> Ana.
> *
> * 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
845-246-0774
*
* 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/
