--- On Sat, 29/8/09, Amanda Agan wrote:
> I am using the "mim" command to do estimation on multiply
> imputed datasets. I did not use "ice" to create these
> datasets as I wanted to use an estimation scheme not
> allowed by ice (desmat). Instead, I created the multiply
> imputed datasets myself and put them together using mijoin
> (as shown here:
> http://www.ats.ucla.edu/stat/Stata/library/ice.htm
> under "other issues").
>
> Then, when I use the "mim" prefix to run a very simple
> regression i get the following error:
>
> " file var/tmp_mim_ests1.ster could not be opened for
> output "
>
> Any thoughts?
My first thought is that both -mim- and -desmat- are user
written program, so per the Statalist FAQ you should
specify where you obtained these. This is important these
programs tend to float around in cyberspace in multiple
locations and in multiple versions.
My second thought is that your manually imputation model
likely to be very problematic. -desmat- is just a
convenience command for creating variables. Manually
creating correct imputations in this case is virtually
impossible if you don't exactly know what -desmat- does
and how to manually reproduce it's behavior. If you know
that already, than it is much much saver to use -ice-,
circumventing -desmat-, and make use of the -passive()-
option in -ice-, as discussed here:
http://www.stata-journal.com/article.html?article=st0067_1
So I would advise to take a look at the articles in which
-desmat- is described (type -help desmat- to get the
references, which are in the STB which are available here:
http://www.stata.com/bookstore/stbj.html ) Than see if you
can recover from those articles how to manually recreate
the variables, and use -ice- with the -passive()- option
instead of your manual imputation model.
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
