 |
 |
Hi pals, I have a question about the way partial correlations were estimated for the HCP megatrawl using FSLnets. Going through the 'ridgep' case in nets_netmats.m, it looks like the covariance matrix for each subject is scaled by the square root of the mean of the variances squared - yes, no, maybe so? from line 88 of nets_netmats.m: grot = cov(<component timeseries>); grot = grot/sqrt(mean(diag(grot).^2)); I'm just trying to figure out why I need to square the variances before I take the mean. Can someone give me a quick explanation or point me towards a good reference? Thanks in advance for your help! mb