Hi Daniel,
you can always specify your own design matrix X in mri_glmfit. So just create those columns, fit and test. Of course you need to make sure that the order of subjects in Y is identical with the one used in X.
Cheers, Martin
On 11/06/2015 06:14 PM, Yang, Daniel wrote:
Dear FreeSurfer experts,
I have a group variable (X) with two levels and I want to test if there is a curvilinear interaction with age (Z). Can this be tested in FreeSurfer?
Specifically, the model would be
Y = b0 + b1X + b2Z + b3Z*Z + b4X*Z + b5X*Z*Z + error
I want to test b5 while controlling for b0-b4.
Many thanks! Daniel _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer