L-stats

From Waisman Brain Imaging Wiki

In LORETA program to run all statistics: 

Statistical non-parametric Mapping
->Voxel-wise LORETA (binary format) 
  ->Paired Samples

We are currently using defaults, simply double click to enter the info file, and hit GO!

[edit]

Info File

Have to set up an external text file in order to run stats

Example file is infoGam.txt

First line of info file: 

NS NT NE Cond1TF1 Cond1TF2 Cond2TF1 Cond2TF2

NS = number of subjects
NT = number of frames (time or discrete frequency (8 bands for 'sknk')
NE = number of voxels (LORETA voxels=2394)
Cond1TF1, Cond1TF2 = time/frequency window limits in range 1,2,3,. (use '3 4' alpha for 'sknk')
Cond2TF1, Cond2TF2 = time/frequency window limits in range 1,2,3,..(use '3 4' alpha for 'sknk')


Following lines of file: 

file name for subject number 1, condition 1
file name for subject number 2, condition 1
file name for subject number 3, condition 1
....
file name for subject number 1, condition 2
file name for subject number 2, condition 2
file name for subject number 3, condition 2

(example W:\exp\skunkwrk\eeg_data\cohort1\sknk201\ICA\ICA_data\NI.lorb).

For skunkwrk we will run 3 tests per analysis method:(Eventually add in MX version) 

PU-NI
PU-RE
RE-NI
(MX-PU)
(MX-NI)
(MX-RE)

Each directory will have 3 analysis paths resulting in 3*3=9 tests total (3*6=18 total)

Retrieved from "http://wiki.brainimaging.waisman.wisc.edu/index.php/L-stats"