Hi Anderson,
Thanks for the suggestions and I'll keep that in mind if this becomes too Freesurfer-centric for posting. I have been running over the weekend but noticed my simulations are still going so I wanted to know if the parallelize toolbox and cluster setup in MATLAB would speed this up or not?
Finally, is this model I have created centering across groups but different slopes or would I have to use the -demean flag to achieve this (so that covariates such as age, education and disease severity are evaluated when centered across all groups as opposed to at the y=0 intercept)? Or would it be easier to enter demeaned covariates for age,Edu,DS if the -demean flag affects EVs 1-6 in an unintended way?
/ContrastName1 Group1 > Group2
/ContrastName2 Group1 < Group2
/ContrastName3 Group1 > Group3
/ContrastName4 Group1 < Group3
/ContrastName5 Group2 > Group3
/ContrastName6 Group2 < Group3
/ContrastName7 M > F
/ContrastName8 M < F
/ContrastName9 Grou1/Group2 - M/F Interaction
/ContrastName10 Group1/Group3 - M/F Interaction
/ContrastName11 Group2/Group3 - M/F Interaction
/NumWaves 9
/NumPoints 11
/Matrix
1 1 -1 -1 0 0 0 0 0
-1 -1 1 1 0 0 0 0 0
1 1 0 0 -1 -1 0 0 0
-1 -1 0 0 1 1 0 0 0
0 0 1 1 -1 -1 0 0 0
0 0 -1 -1 1 1 0 0 0
1 -1 1 -1 1 -1 0 0 0
-1 1 -1 1 -1 1 0 0 0
1 -1 -1 1 0 0 0 0 0
1 -1 0 0 -1 1 0 0 0
0 0 1 -1 -1 1 0 0 0
EV1: Group1-M
EV2: Group1-F
EV3: Group2-M
EV4: Group2-F
EV5: Group3-M
EV6: Group3-F
EV7: Age
EV8: Education
EV9: Disease severity
/NumWaves 9
/NumPoints 149
/Matrix
0 1 0 0 0 0 70 13 0
0 1 0 0 0 0 63 16 0
0 1 0 0 0 0 64 21 1
0 1 0 0 0 0 65 20 2
1 0 0 0 0 0 80 20 0
1 0 0 0 0 0 45 16 0
1 0 0 0 0 0 57 14 2
1 0 0 0 0 0 70 17 0
1 0 0 0 0 0 71 17 0
.
.
.
Thanks,
Ajay
|