On 29 Jan 2004 07:43:37 -0800, [EMAIL PROTECTED] (Nuage) wrote:

> Hello,
> 
> I wanted to know how to perform a MANOVA(Multivariate Analysis of Variance).
> Anyone could help me?
> Is there a special way to do it, compared to an ANOVA?

One simple, reduced case of MANOVA is the ordinary 
discriminant function; or, bigger, the discriminant function  
with multiple groups.   You can look at the *general* problem
as the eigenvector problem that is solved in canonical 
correlation, so there are multiple variables on each side
of an equation.

One other reduced case is  the OLS-multiple regression;
the OLS-MR   is the same simple matrix as the oneway ANOVA.

I would not try any of these by hand, these days.

-- 
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
"Taxes are the price we pay for civilization." 
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to