-------- Original Message --------
Subject: categorical variables in pls
Date: Wed, 25 Apr 2012 16:46:51 -0400
From: Rodrigo Lima <rodrigo.l...@mail.mcgill.ca>
To: morphmet@morphometrics.org <morphmet@morphometrics.org>
Dear morphometricians,
Does it make sense to use categorical variables in a PLS analysis?
I have landmark configurations of animals collected in 3 ecozones, and
these ecozones succeed each other in a (more or less) latitudinal
gradient. I think it doesn't make sense to give them values 1, 2, and 3,
but would it make sense to code them as dummy variables and use these in
PLS together with other "environment" variables?
Thank you,
Rodrigo
