29 Jan
2010
29 Jan
'10
3:01 p.m.
FDR is NOT more conservative than cluster-based methods, in general. The smoother the data, the more conservative FDR is.
In my experience with surface-based data, FDR has been less sensitive than cluster-based methods -- perhaps because my data was very smooth.
Stefan Brauns wrote:
We are well aware of the fact that FDR and Monte Carlo simulation are very different statistical methods and that FDR is more conservative.