[S] Generating Multivariate Cauchy

andrey@utstat.toronto.edu
Mon, 23 Nov 98 16:45 EST


Oops! Sorry about a small error in my posting.
The solution ``rnorm(n)/rnorm(1)'' assumes a particular
`covariation' structure for the multivariate distribution.
Usually the numerator needs to be linearly transformed
to introduce the `covariation' desired.

A.F.

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news