Peter D. Wilson (peter at dflora.com) wrote:
: Posting for a friend:
: I have been asked to trace some software which performs
: "Trended Correspondence Analysis". It is to be used in botany and is
: believed to have been created by a botanist. Any help would be
: appreciated.
: - peter
: --
: peter d. wilson
: email: peter at dflora.com (NeXTMail OK)
: Voice: (403) 283-9743 Fax: (403) 283-0036
: Independent NextStep Developer
Before using Detrended Correspondence Analysis, you might
want to read
James, FC and CE McCulloch, 1990, Multivariate analysis in ecology and
systematics: panacea or Pandora's box? Annual Reviews of Ecology and
Systematics 21:129-166.
For what it's worth, I recently looked at DCA and some other multivariate
analyses for a particular ecological use, and decided to use the good ol'
principal components (PCA) and factor analysis after all. DCA is intended
to improve upon the PCA, but carries with it some new problems (see James
and McCulloch). Given
that neither technique is a 'magic bullet,' it might be just as well to
stick with PCA, as it is more familiar, and most stat programs do it
(SAS, SPSS, at least). DCA might (or might not) be discarded in ten
years, but PCA has a historical tradition and straightforward mathematics
(just pulling out eigenvalues) which make it a likely candidate for
durability.
Regardless of what you or, actually, your friend uses, don't expect
miracles from any multivariate
technique. It will help you organize your data, but you are still going
to need to wrack your brain to decide what it tells you that is useful,
if anything at all.
--
Jim Gilliam
james_gilliam at ncsu.edu (preferred) or jfgzo at unity.ncsu.edu
Department of Zoology
North Carolina State University
Raleigh, NC 27695-7617
