Print

Print
Hi,

I'm dealing with a similar model of my own, and was wondering why you didn't suggest adding a mean regressor to model the intercept, as well. From what I've seen in earlier posts, an additional mean regressor column of all 1s is needed in models like this (modeling a continuous variable's effect on the whole group).

Thanks,
Amanda

On Thu, Mar 29, 2012 at 5:59 AM, Stephen Smith <[log in to unmask]> wrote:
Yes
Cheers


On 28 Mar 2012, at 05:09, Xu Chen wrote:

Dear FSLers

  I am using TBSS to explore if there is any quadratic effect of age on my FA.  I set up a design matrix of 2 EVs (EV1=demeaned (age^2), EV2=demeaned age) with the contrasts of [1,0] &[-1,0].    Is this a correct configuration? Thanks

Jerry


---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------