Hi - highpass filtering is applied to the model and the data before
running the timeseries modelling; the timeseries modelling includes
the whitening step, which is applied to the pre-filtered data and model.
Cheers.
On 23 Oct 2007, at 15:42, Brian Lenoski wrote:
> Hello,
> I have a similar question along these lines. When doing pre-
> whitening is the temporal filter applied before or after the model
> is multiplied by the pre-whitening matrix? I would guess that the
> pre-whitening would be performed after temporal filtering because
> the autocorrelation structure has been estimated from the
> temporally filtered model?
>
> Best,
> Brian
>
>
> On 10/23/07, Mark Jenkinson <[log in to unmask]> wrote: Hi,
>
> The motion correction parameters are filtered by the temporal
> filter in
> the same way
> that both the data and the rest of the model is. This is done so that
> the data and model
> are well matched, since if y=X*beta + noise before filtering, then you
> need to filter
> both left hand side and right hand side (y and X) to maintain the
> equation.
>
> Hope that helps.
> All the best,
> Mark
>
>
>
> Guido Biele wrote:
>
> >hi,
> >when looking at the motion correction paramters included in the
> design matrix by feat it apppeared
> >to me that they differ from the original parameters generated by
> mcflirt.
> >it seems that the the values in the design matrix are detrended
> (to remove autocorrelation?).
> >
> >how are the raw motion correction parameters modified before they
> enter the model?
> >could anyboy please explain briefly or refer to the appropriate
> reference?
> >
> >cheers - guido
> >
> >
>
------------------------------------------------------------------------
---
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
------------------------------------------------------------------------
---
|