Basically x= randn(100,15)
x_centered=(eye(100)-(1/100)*ones(100,100))*x to subtract the mean to center the data
[s,d,v]=svd(x_centered);
I need to find the principal components
which just the eigenvalues but according to my professor {divi}for i = 1 to 15.
Then I am asked what is the pca with 3, 4, 5 components. For 3 components is it just [d1v1 d2v2 d3v3]
Then I am asked to find the mse of it. I am confused what do I compare the PCA of 3 components to to take the F- norm?