If i’m running linear regression analyses for each of the 400 ROIs across the whole brain, is there a better alternative for correcting for multiple comparisons than using bonferroni or FDR?
For limited numbers of tests, Bonferroni should not be too bad. FDR is fine. You can also use permutation tests, but I don’t think that this is worth the effort.
Best,
Bertrand
The permutation test sounds interesting. Do you have a reference for it?
Well. There’s nothing special about these tests. Maybe
Best,
Bertrand