|
> I used ttest2 on two 3D matrices. I would like to make a t distributions
> of the t scores of each voxel and display the ones that are significant at
> alpha=0.05. Both matrices are equal...
Sylvie, I am not sure I understand the question, but here are a few lines of
code that try to detect ways in which two random things are significantly
different:
s = rand; % some random standard deviation
a1 = s*randn(10,10,100); % random background noise
a2 = s*randn(10,10,100); % another
a1(1:5,1:6,:) = a1(1:5,1:6,:) + s/2; % signal equals 0.5 * std dev
h = ttest2(a1,a2,0.05,'both','equal',3)
-- Tom
|