Post-estimation after mi estimate: mimrgns error

Welcome to the forum for runmlwin users. Feel free to post your question about runmlwin 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 to runmlwin: Running MLwiN from within Stata >> http://www.bristol.ac.uk/cmm/software/runmlwin/
Post Reply
amarkovitz
Posts: 7
Joined: Mon Aug 22, 2016 10:19 pm

Post-estimation after mi estimate: mimrgns error

Post by amarkovitz »

Hello all,

I have used the very helpful posts here to run runmlwin after multiple imputation using mi estimate. However, I need to make out-of-sample predictions after this procedure and have found that mimgrns from SSC is one of the few (or only?) possible way to do this despite some caveats to the standard errors produced, which I'm aware of. However, running mimrgns after runmlwin produces the following error:
"variable var not found
an error occurred when mi estimate executed mimrgns_estimate on m=1"

Does anyone know of tricks to use runmlwin and mimrgns in conjunction or of another way to do out-of-sample prediction after mi estimate + runmlwin? It seems that doing it by hand and applying Rubin's rules would be prohibitively difficult, especially in the case of my rather complicated dataset.
amarkovitz
Posts: 7
Joined: Mon Aug 22, 2016 10:19 pm

Re: Post-estimation after mi estimate: mimrgns error

Post by amarkovitz »

To close out this topic (unless anyone has advice), I believe the following post answers my question about whether mimargins can be used in conjunction with runmlwin. The answer is no.
https://www.cmm.bristol.ac.uk/forum/vie ... edb4#p4431

It sounds like I would have to do it using first principles which is too time consuming for this project or I can use the much slower built in Stata multilevel modeling commands in conjunction with mimgrns.
Post Reply