On Tue, 12 Jan 2010, Lars PĂ„lsson-Syll wrote:
I can see that Cholesky decomposition works perfectly in the
Monte Carlo method for simulating systems with multiple
correlated variables. But I'm not quite sure WHY, so if you have
any readable references I would be more than happy to take a
look at it.
There may be better discussions but this one seems quite clear:
http://comisef.wikidot.com/tutorial:correlation
(By the way, I couldn't find anything in the Gretl-manual about
the operator *= that was used in the script).
We support the "@=" shorthand form for all the common operators.
I suppose it should be in the manual if it's not there.
Allin