It suspect it does regression omitting variables with missing values (as
part of the best subset of variables regression approach). Observations
with different missing value patterns are imputed in different steps, and
for each
step different variables need be thrown out because of missing values.
Matt
-----Original Message-----
From: Zhehui Luo [mailto:[email protected]]
Sent: Wednesday, January 22, 2003 10:03 PM
To: [email protected]
Subject: st: command impute
Hi,
There is a command -- impute depvar varlist.
Who knows how exactly the impute command predict the missing depvar when
some of the variables in varlist are missing. The reference guide is not
very clear on this issue. I thought the predict command after regression
will give missing values for the predicted depvar whenever there is missing
in any of the regressors. So how on earth does impute make the prediction
when the regressors have missing values?
Thanks!
Zhehui
*
* 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/
