Simulating Correlated non-Normal Random Variables

*To*: mathgroup at smc.vnet.net*Subject*: [mg32416] Simulating Correlated non-Normal Random Variables*From*: "Coleman, Mark" <mark.coleman at dri-wefa.com>*Date*: Sat, 19 Jan 2002 01:17:14 -0500 (EST)*Sender*: owner-wri-mathgroup at wolfram.com

Greetings, This is not a Mathematica question per se, but I'm hoping members of this list might be able to point me in the right direction. I'm constructing a Monte Carlo simulation procedure, where I need to draw a n-vector of correlated random variables, given that I know the univariate distribution of each element of the vector and the corresponding covariance matrix. This problem is straightforward for normally distributed random variables, where one can create N(0,1) variates and multiply by the matrix square root of the covariance matrix to created the correlated sample. I am dealing with non-normally distrubuted random variables, however. In my case they have the Beta distribution. Can someone point me to a way to draw an n-vector of correlated beta random variables? Thanks, -Mark