imputed dataset (long form) & realcomImpute for Linux

Welcome to the forum for REALCOM users. Feel free to post your question about REALCOM here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

Go REALCOM (Developing multilevel models for REAListically COMplex social science data) >> http://www.bristol.ac.uk/cmm/software/realcom/
Post Reply
gindotampu
Posts: 7
Joined: Tue Jun 05, 2012 1:55 am

imputed dataset (long form) & realcomImpute for Linux

Post by gindotampu »

Dear colleagues,
two topics really: One, is there an option to set the output dataset (imputed dataset) as say ten completed dataset instead of one dataset with ten additional set of imputed variables? In other words the so called long form instead of wide form? I want to do separate analysis on each completed dataset in other software [at least that was the principle behind Rubin's multiple imputation if I recall].
Two, I use realcomImpute for Stata with Stata of course and Matlab runtime exec in Windows operating system. Is there an equivalent set-up for Linux (Ubuntu)? Or, I've got Stata for Linux, what else do I need to use realcomImpute there, quite apart from the Wine-MLwiN route?

Many thanks for a fantastic software,
Gindo
PS In case people ask, we haven't got the 64bit Windows OS but I have 64bit Linux OS & Stata.
ChrisCharlton
Posts: 1248
Joined: Mon Oct 19, 2009 10:34 am

Re: imputed dataset (long form) & realcomImpute for Linux

Post by ChrisCharlton »

1) There isn't an option to save as one output dataset, however as you are using Stata you should be able to use Stata's http://www.stata.com/help.cgi?mi commands to convert the data to the format that you want. To do this you would use:

Code: Select all

mi import flongsep ...
to load all the data files and then:

Code: Select all

mi convert wide ...
to create a single wide version of the data. You can then merge in your original data and use the command:

Code: Select all

mi convert flongsep ...
to save the data as separate files again.

2) As far as I know the -realcomImpute- command is written entirely in Stata code, so should work on any operating system supported by Stata. We do not currently provide a version of Realcom-Impute that has been compiled using the Linux version of Matlab, so to use this you would have to install the Matlab runtime and Realcom-Impute under Wine.
gindotampu
Posts: 7
Joined: Tue Jun 05, 2012 1:55 am

Re: imputed dataset (long form) & realcomImpute for Linux

Post by gindotampu »

Thanks very much Chris for the prompt reply.

On the Linux question: since the university has licence for Matlab for Linux, is there a RealcomImpute binary object that can be put on Linux? (Callable from Stata somehow).
Cheers,
Gindo
University of Manchester
ChrisCharlton
Posts: 1248
Joined: Mon Oct 19, 2009 10:34 am

Re: imputed dataset (long form) & realcomImpute for Linux

Post by ChrisCharlton »

Unfortunately this currently isn't available and we don't have access to a Linux version of Matlab to build it. The Realcom-impute application isn't directly called from Stata it just outputs files that the -realcomImpute- Stata command can use. This means that you can impute the data using a Windows machine/Wine and then use the Linux version of Stata to load and analyse it. I'm sorry I can't give you a better solution than that.
gindotampu
Posts: 7
Joined: Tue Jun 05, 2012 1:55 am

Re: imputed dataset (long form) & realcomImpute for Linux

Post by gindotampu »

Dear Chris,
Thanks again for the prompt reply. It clears my options.
Again, many thanks for a fantastic software.
Gindo
Post Reply