Hi - the -1 option is for the case where the model is just a single
group mean -i.e. one EV with all 1s in it.
To do a single regression just put the regression values in a single
EV, making sure that the EV is demeaned (ie you need to remove the
average of all values so that the EV has zero mean). Then use the -D
flag to tell randomise to also demean the data, and all should then
be straightforward.
Cheers.
On 24 Aug 2007, at 06:19, Nikki Horne wrote:
> Dear FSL users,
>
> I am pretty new to FSL. I am trying to navigate the new release,
> which looks
> very promising, and I have questions regarding how to perform a simple
> one-sample regression on TBSS data in randomise (the examples in the
> randomise manual do not include one for regression).
>
> 1. What is the basic command needed and how do I set this up in
> glm? I read
> in the manual that the -1 option should be used for a single group of
> subjects, but I'm not sure what this will look like.
>
> 2. Do I set up two EVs, one with all 1s and the other with the data
> to be
> regressed (I tried this approach, but I got a warning that my EV's
> were
> highly correlated)?
>
> 3. Do I need to demean the regressor data, use the -D flag, or both
> in this
> case?
>
> Thank you for your time and assistance,
>
> Nikki Horne
------------------------------------------------------------------------
---
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
------------------------------------------------------------------------
---
|