[R] multiv. lm

Prof Brian Ripley ripley at stats.ox.ac.uk
Mon Jun 28 14:34:59 CEST 1999

> Date: Mon, 28 Jun 1999 13:46:09 +0200 (MET DST)
> From: Torsten Hothorn <hothorn at amadeus.statistik.uni-dortmund.de>

> is there a basic function for multivariate linear models  (as basic as 
> "lm" is for the univariate case) and can one derive MANOVA and other methods 
> from it?  I found "mlm.R" but there is only a summary method defined. 
> If not, is somebody working on this problem, otherwise I would refresh some
> older code and include multivariate testing (Roy, tests based on
> linear scores etc.pp.),

Well, mlm and aov are the basic methods, and one can derive MANOVA etc
from them, as the manova() function in S-PLUS (but not S nor R) does just
that (calls aov, makes the class of the result "manova").  The real
work is done in summary.manova to produce the tests, which are really
the only difference between multiple and multivariate linear models.
(Summary.manova cheats by using approximate distribution theory.)

As far as I know no one is working on this, and we would welcome 
contributions.  One reason it is not there is that I (who contributed the 
mlm and aov code) have never seen a problem needing multivariate testing
(which is not to say that they do not exist, just are low priority for me).

Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list