Le us create N independent and identitcally distributed (iid) normal random vectors with zero mean and specific covariance matrix. Each vector is of size p. Let $\Sigma$ be the specific covariance matrix.

    p = 20;
    N = 50;
    M = chol($\Sigma$)
    Matr =  M' * randn(p, N) 

In fact I want to know why in all references, they use chol?. Why they don't simply write Matr = randn(N, p) * ($\Sigma$) ? So what is the advantage of chol (or cholcov) in this case?

I made some results, and I noticed that they significantly differ when I remove the chol !! Even it takes more time for the computation..why??

Any help will be very appreciated!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.