 |
 |
Hi Moran, say you have two groups, 3 subjects in each (1:3 in group1, and 4:6 in group2), a sample permutation is using the same data, BUT grouping them as 1,3,4 in group1 and 2,5,6 in group2 (changing the design matrix, which is alright under H0, which is sth like "no difference across groups") The number o possible permutations then depends on the number of possible label-exchanges in the design matrix, In your case, the answer to the question: "what is the number of possible ways to divide 6 subjects into two groups, while having _one_ subject in group1 and _five_ subjects in group2" which is identical to the possible number of permutations, is (unfortunately) 6 :( and that's why ... cheers 2009/3/8 Moran Artzi <[log in to unmask]> > Hi, > > I have few questions regarding to VBM data analysis: > > 1. After running fslvbm_3_proc the output are the smoothed files > GM_mod_merg_s2-4 well understood, and the GM_mod_merg_s2_tstat1.nii.gz > (attached) I do not understand how these tstats maps can help me decide > which smoothing is the most relevant to feed into a full run of randomise, > and > which threshold to use for the cluster-based thresholding. > > 2. On the next step, running the follow randomise script (in order to > compare > between the 2 groups) > moran@tux:~/Moran/ALS/VBMn/template_list/stats$ > randomise -i GM_mod_merg_s3 -m GM_mask -o fslvbm -d design.mat -t > design.con -c 2.3 -n 5000 –V > (Attached - design file- > only for the test analasis, 2 groups with 3 subject each) > > The following output appeared: > Loading Data: ****** > Data loaded > 6 permutations required for exhaustive test of t-test 1 > Doing all 6 unique permutations > Starting permutation 1 (Unpermuted data) > Starting permutation 2 > Starting permutation 3 > Starting permutation 4 > Starting permutation 5 > Starting permutation 6 > > For each one of the 4 t-test. > > Why 6 and not 5000 permutation for each test? and what dose its mean > Unpermuted data mean? > In addition – the all step long was only about 1 mint. > Did I do something wrong? > > Many thanks > Moran > > > -- G. Salimi-Khorshidi, D.Phil. Student, Dept. of Clinical Neurology, University of Oxford. [log in to unmask] http://www.fmrib.ox.ac.uk/~reza FMRIB Centre, John Radcliffe Hospital, Headington, Oxford, OX3 9DU Tel: +44 (0) 1865 222466 Fax: +44 (0)1865 222717