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: post-estimation with MI data
From
[email protected] (Yulia Marchenko, StataCorp LP)
To
"[email protected]" <[email protected]>
Subject
Re: st: post-estimation with MI data
Date
Mon, 10 Jun 2013 10:17:05 -0500
Ryan Wells <[email protected]> asks whether there is an equivalent to
-e(sample)- after the -mi estimate- command:
> What is the equivalent of specifying 'if e(sample)' after running a command
> like:
> mi estimate, post: svy, subpop(subsample): logit `lhs' `rhs'
>
> since e(sample) does not appear in the ereturn results? Is there a way to
> get the equivalent info from MI estimation results?
Generally, there is no analog of -e(sample)- for MI estimation results because
estimation is being performed on multiple datasets and can potentially use
different sets of observations.
You can store estimation samples from each individual dataset by using the
-esample()- option of -mi estimate-. You must be in the -flong- or -flongsep-
MI style to be able to store estimation samples.
-- Yulia
[email protected]
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/