Hello

Sorry for reposting again.

Would really appreciate a feedback on the following mail.

Thanks in advance for your time

Raka

----- Forwarded Message ----
From: raka maitra <rakamaitra@yahoo.com>
To: Douglas N Greve <greve@nmr.mgh.harvard.edu>
Sent: Fri, January 15, 2010 9:34:02 PM
Subject: Re: [Freesurfer] manually specifying design matrix

Thank you Doug.

I am trying to create a design matrix for repeated measures anova.

Lets say subjects A and B have been measured at sitting and standing positions.

As the same subjects have been measured at the same positions, hence both the factors i.e subjects and positions are fixed factors.

Hence a design matrix exploring the within subject effects of positions would be

1 0 1 0

0 1 1 0

1 0 0 1

0 1 0 1

The first two columns are position and the last two are the subjects.

But mri_glmfit does not accept this design and I get the following error message:

'Matrix condition is 1e+08

ERROR: matrix is ill-conditioned or badly scaled, condno = 1e+08

Possible problem with experimental design:

Check for duplicate entries and/or lack of range of

continuous variables within a class.'

Looking forward to your suggestion

Thanks in advance

Raka

From: Douglas N Greve <greve@nmr.mgh.harvard.edu>
To: raka maitra <rakamaitra@yahoo.com>
Cc: freesurfer@nmr.mgh.harvard.edu
Sent: Thu, January 14, 2010 7:05:28 PM
Subject: Re: [Freesurfer] manually specifying design matrix

If you specify it manually, you don't specify groups, etc, or pass an
fsgd file. You just create your design matrix pass it to mri_glmfit
directly with the --X flag.

doug

raka maitra wrote:
>
> Hi all
>
> I want to specify a design matrix manually.
>
> In the usual pipeline for analysis in freesurfer, one specifies the
> groups, regressors in the fsgd and then when mri_glmfit is run, it
> creates the design matrix.
> For eg if I have two classes with 3 subjects each, in my fsgd,
> mri_glmfit would create the following design matrix
> 1 0
> 1 0
> 1 0
> 0 1
> 0 1
> 0 1
>
> If I want to specify the above design matrix manually, how and where
> do I specify the groups, regressors ?
>
>
> Thanks in advance
> Raka
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Freesurfer mailing list
> Freesurfer@nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

--
Douglas N. Greve, Ph.D.
MGH-NMR Center
greve@nmr.mgh.harvard.edu
Phone Number: 617-724-2358
Fax: 617-726-7422

Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html

_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer