Re: joint distribution of correlated uniform
From: Herman Rubin (hrubin_at_odds.stat.purdue.edu)
Date: 08/21/04
- Next message: Richard Harter: "Re: Finding Percentiles WITHOUT Sorting"
- Previous message: Anonymous: "Re: Finding Percentiles WITHOUT Sorting"
- In reply to: Yan Xie: "joint distribution of correlated uniform"
- Next in thread: Richard Ulrich: "Re: joint distribution of correlated uniform"
- Messages sorted by: [ date ] [ thread ]
Date: 21 Aug 2004 08:25:49 -0500
In article <tzb0eoc72euz@legacy>, Yan Xie <yxie2@uky.edu> wrote:
>Hello, all,
>1. Is it possible to get the joint distribution of n correlated
>uniform(0,1) random variables given the correlation matrix? If yes,
>any suggestion for it?
>2. Dose anybody have or happen to know an algorithm to generate
>positive definite matrix?
>Thanks in advance.
It is not possible even for normal unless it is assumed
that they are jointly normal; there is a "natural"
multivariate normal distribution which can be
characterized in many ways, one being that every linear
combination is normal.
As the normal distribution is the only nontrivial one
with a finite variance for which the distribution of
the sum of two independent random variables with that
distribution has a distribution of the same form, a
natural generalization to uniform does not exist.
-- 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, Department of Statistics, Purdue University hrubin@stat.purdue.edu Phone: (765)494-6054 FAX: (765)494-0558
- Next message: Richard Harter: "Re: Finding Percentiles WITHOUT Sorting"
- Previous message: Anonymous: "Re: Finding Percentiles WITHOUT Sorting"
- In reply to: Yan Xie: "joint distribution of correlated uniform"
- Next in thread: Richard Ulrich: "Re: joint distribution of correlated uniform"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|