Hi Bennett,
> I am interested in comparing the effects of preprocessing. I am
> exploring two analysis method variants that produce different
> statistical results. Visually, the residual mean square is much higher
> for one method than the other. I would appreciate advice as to how to
> quantitatively compare the residuals, especially how to properly
> correct for the multiple comparisons/resels. Please let me know if
> there is better way of assessing possible reductions variability of
> one method relative to another.
>
I'm not sure that it will do exactly want you want but Tom Nichols' SPMd
toolbox (Statistical Parametric Mapping Diagnosis & Exploration) should
be useful for looking at distributions and other stuff like that.
http://www.sph.umich.edu/~nichols/
Best,
- cyril
|