In article <hmm_3.3961$[EMAIL PROTECTED]>,
Attila <[EMAIL PROTECTED]> wrote:
>Hello All!
>I am looking for a good ( and understandable to a non-mathematician ) source
>that explains the bivariate normal distribution or ( if there is any ) a
>multi ( 3 variable )
>equivalent. Namely, I would like find a method to sample a 2 or 3 variable
>normal distribution
>with a given mean and std. dev. like it is possible to do for a 1 variable
>with, say Tukey-Lambda...
There are lots of ways of characterizing a multivariate
normal random variable. It can be done in terms of
densities (if they are linearly independent) or in
terms of the characteristic function or moment generating
function in any case.
An easy to understand characterization, which requires
less mathematics, is that a set of random variables are
multivariate normal if and only if they are linear
combinations of a set of INDEPENDENT normal random
variables. This also answers the second question; there
are better ways of getting univariate normal variables.
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
[EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558
