mr.mgh.harvard.edu<mailto:freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] Multivariable linear regression with binary variables
You'd have 512 groups total, which is probably way too many regardless
of whether you do so manually or automatically. You need to think about
how you want
November 2018 18:03:16
To: freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] Multivariable linear regression with binary variables
You'd have 512 groups total, which is probably way too many regardless
of whether you do so manually or automatically. You need to think about
how you wa
You'd have 512 groups total, which is probably way too many regardless
of whether you do so manually or automatically. You need to think about
how you want to model all those variables since modeling all possible
interactions is not going to be possible.
On 11/08/2018 12:45 PM, Worker, Amanda w
External Email - Use Caution
Hi,
I am trying to run a multivariable GLM with mri_glmfit and I am currently
setting up my fsgd file. I have 9 binary variables and 1 continuous variable.
My understanding is that in the fsgd file, my "class" would be every possible
combination of
