Radim,

Don't know if you have your answer yet, but you just need some basis, not
the best basis.  Thus, you can do block-wise decomposition of Y and you
nearly have it.  You still have to divide by the number of blocks so that Q'
Q = I, but other than that, I think you are set.

2010/9/7 RadimRehurek <[email protected]>

> But I don't understand the step where you orthonormalize Q: it looks like
> you want to do it in blocks of Y, but orth(Y1 ; Y2 ; Y3 ...) != orth(Y1) ;
> orth(Y2) ; orth(Y3) ... (the first two equations of sd.pdf). What is the
> trick behind that?

Reply via email to