Hi - those are the different methods for multivariate discriminant analysis - all simple methods. See the matlab code in FSLnets for more information about what the different methods are. The percentages are overall accuracy when running a leave-one-out test. These results are rather close to 50% (chance) - so not very hopeful wrt discriminating between your two groups! Cheers. On 10 Nov 2012, at 16:16, Cornelius Werner wrote: > Hello list, dear experts, > > I am currently analyzing a set of data comprising resting state fMRI measurements from both 23 patients and 23 controls by means of dual regression. I am also having a look at the resulting dr1 files with FSLNets, which yields interesting results. However, I am a little dumbfounded on how to intepret the output given by > > [lda_percentages]=nets_lda(netmat2,23) > > lda_percentages = > > Columns 1 through 6 > > 58.6957 60.8696 63.0435 41.3043 43.4783 65.2174 > > Column 7 > > 56.5217 > > What do these percentages represent? Is this in any way a significant result? Are they the results of the lda(1-7,n) calls at the end of the nets_lda.m script? Where could I read up on the method employed? > > OK, thanks a lot for helping me! > Cheers, Cornelius > > -- > Cornelius Werner > cornelius.werner<at>gmail.com --------------------------------------------------------------------------- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ---------------------------------------------------------------------------