Re: Uncorrelated random variables

[Steve C. Thompson]

On 29 Jun 06 20:03PM, Evgeny Turchyn wrote:
> Suppose that we use Octave to generate a sample of
> size c from, say, normal distribution using function
> normal_rnd (0, 1, 1, c). Can we consider the obtained
> vector as a realization of vector which elements are
> uncorrelated random variables with distribution
> N(0;1) ? (The important moment for me is
> uncorrelatedness.)

GNU Octave uses a state-of-the-art random number
generator.  Though it is impossible to generate purely
random, uncorrelated random variables on a
deterministic machine like a digital computer, the
algorithms used by GNU Octave result in randomness
exceeding the requirements for most conceivable
applications.

Others will correct me if I'm wrong (I have only a
vague understanding of the exact algorithms used), but
I believe Octave uses the Mersenne Twister random
number generator to generate uniform random variables,
and the Marsalia-Tsang ziggurat method for generating
normal random variables.

Steve