Closed danielrmrtns closed 8 months ago
Hi Daniel,
This can happen with unusual data (like you seem to have indeed, binarized) and/or when too few permutations are done.
The Gamma approximation tends to be more robust in these cases ("-approx gamma"). I'd give it a try.
All the best,
Anderson
Dear PALM team,
I've been having some troubles with the usage of -accel tail. Sometimes, depending on the input data, the p-value maps are not calculated and the following error gets caught:
Without -accel tail, this does not arise. Do you reckon what could be causing this?
Note: The design is of a standard 2-sample unpaired t-test, with n=30 binarized volumes in the 1st group against n=79 in the 2nd.