Hi guruji,
I won't bore you with the details about how I got myself into this mess, but I'm wondering if there is a bootstrapping procedure written in AFN-ese? Given the variety of things that require bootstrapping, I wouldn't necessarily expect a full bootstrapping suite or anything, but perhaps a command that could be inserted into a shell script that does the hard part of randomly sampling with replacement from:
- two n-length 1D arrays (easy).
- an n-length 1D array and n 3D datasets (e.g., doing a voxel-wise correlation against a single behavioral variable).
- n 3D datasets and another n 3D datasets (e.g., correlating T1 values against response coefficients, voxel-wise).
I appreciate that this is unwieldy and potentially gross--just checking before I start to port things into MATLAB as to whether a sufficient need for bootstrapping has emerged in the AFNI community.
Thanks!
Paul