Hi,
> I get the following error when running estimate in MarsBar. Any suggestions
> will be deeply appreciated.
> ReML Iteration : 24 ...1.134142e+002
> ReML Iteration : 25 ...-3.030891e+000
> Initialising parameters : ...computing???
> Index exceeds matrix dimensions.
Yes, this is probably because the ReML is failing to find a good match
to your covariance structure, therefore generating a rank deficient
estimate of the covariance, and this ends up generating - er - the
error you see. Try switching your covariance estimate from ReML to
the fmristat method - see:
http://marsbar.sourceforge.net/faq.html#fmristat
Let me know if that doesn't work...
Best,
Matthew
|