Hi,
The tutorial on SEM computes the residual error variance values for the ROI's using the singular values obtained by the SVD of the matrix A (n X T n: number of subjects, T: number of time points, for each ROI).
I would like to know if there is an alternative method to compute the residual error variance values, as I am looking at the data, and applying SEM on only one subject at a time.
Thanks.