 |
 |
Thank you, having the patch would be helpful. I’ve removed duplicate regressors from the matrix, so I’m assuming rank deficiency is coming either from the large number of regressors or from collinearity hidden somewhere in the data (though I think the former is the likely culprit). Is there a command-line way of orthogonalizing the design matrix, or should I try the GUI? In the end, since these are all nuisance regressors, I’m not sure it actually matters if there is some collinearity amongst the regressors (but please correct me if I’m wrong!), and I’m not sure I can easily reduce the number of spike regressors in any principled way. Thanks for any help