From: Paige Miller <paige.miller@kodak.com>
Subject: Re: covariance matrix
Date: Wed, 05 Jan 2000 10:24:21 -0500
Newsgroups: sci.stat.math,sci.math.num-analysis,fj.comp.statistics,gov.us.topic.statistic.reports
Summary: What is a covariance matrix?

Cedric wrote:
> 
> Could somebody explain to me, what is a covariance matrix?

A covariance matrix exists when you have 2 or more variables, and you
want to see how they co-vary. Thus, if you have p variables, you could
compute a pxp (symmetric) covariance matrix. (In the trivial case when
you have 1 variable, the ordinary covariance is a 1x1 covariance
matrix...)

In a covariance matrix, the numbers on the diagonal are the variance of
each variable. The numbers off the diagonal indicate the covariance of
the two variables indicated by the position in the covariance matrix.
The larger the absolute value of the covariance, the greater the
covariation of the two variables. A covariance of zero indicates no
covariance. 

-- 
Paige Miller
Eastman Kodak Company
paige.miller@kodak.com
"It's nothing until I call it!" -- Bill Klem, NL Umpire