 |
 |
Hi, It depends on what kind of ICA analysis you are doing and where you are looking. They are likely to either be based on an alternative hypothesis test or a post-hoc model fit to component timecourses. Thresholding and reporting statistics is not often done for ICA, except in the case of dual regression, where the outputs are actually from randomise. See the wiki and ICA papers for more information. All the best, Mark On 4 May 2014, at 15:22, Xiangning Lyu <[log in to unmask]> wrote: > Hi, > > Can any one kindly help me with the post-stats procedure in ICA? I have seen the results, but I don't quite understand the meaning of p-value and Z-value. What do they mean? And, if I am to threshold the Z-value up to 2.3, what should I do? In the mixture model map, the Z-value seems to go from 0.1 to 1. > > Thanks previously. > Cheers, > Sharon