Hi Tamara and Dr. McLaren,
I am glad there is this simple approach reading SPM.mat that will estimate smoothness - I was having the same problem recently.
I have tried the ResMS approach before, now also with the [n df] = [120*3 303]. This method has given me number in the thousand ranges. I am also not sure what is happening.
Previously I estimated smoothing by rerunning and saving all the ResI file as indicated here
http://akiraoconnor.org/2010/03/24/keeping-residuals-after-spm-estimation/
and used these files with the scripts here to estimate smoothness
http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1505&L=spm&P=R73732&1=spm&9=A&I=-3&J=on&X=C401F8764D658F3231
The results seemed correct. But if there is a simple way to read the smoothness estimates from SPM, I think that saves a lot of time and space.
Thanks
|