External Email - Use Caution I see. Hence I cannot set the contrast of interaction between two covariates if I do not create the interaction term by myself, right?
On Tue, 27 Aug 2019 at 10:53, Greve, Douglas N.,Ph.D. < dgr...@mgh.harvard.edu> wrote: > > Actually, the wiki page is not quite right. For a simple interaction > between age and weight, you would need to create a new regressor = > age*weight, then set the contrast to [0 0 0 0 0 0 1]. If you want an > interaction between group, age, and weight, then you would need two new > regressors, age*weight for group1 and age*weight for group2, then the > contrast will be [0 0 0 0 0 0 1 -1]. I would suggest demeaning both age and > weight prior to computing the product. > > > On 8/27/2019 10:12 AM, Zhi Li wrote: > > External Email - Use Caution > Dear Freesurfer experts, > > May I ask how to understand the contrast of interaction between covariates > in the example https://surfer.nmr.mgh.harvard.edu/fswiki/Fsgdf2G2V? The > interaction of group x age x weight is: > > [0 0 1 -1 0 0 > 0 0 0 0 1 -1] > > Why the interaction between group and weight should be put in the second > row (what does each row mean in the F-contrast matrix)? > > > > If there are three levels in the factor one (three groups), I guess the > interaction of group x age x weight should be: > > [0 0 0 1 -1 0 0 0 0 > 0 0 0 0 1 -1 0 0 0 > 0 0 0 0 0 0 1 -1 0 > 0 0 0 0 0 0 0 1 -1] > > Is this correct? If not how should I set this contrast? > > In addition, if I would like to know the interaction between age and > weight without group, how should I set the contrast then? > > Looking forward to your kind suggestions. > > Thanks and best wishes, > > Lizhi > > > > _______________________________________________ > Freesurfer mailing > listfreesur...@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
