[R] LDA fuction

[Gavin Simpson]

On Mon, 2010-10-11 at 10:18 -0700, vascomc wrote:
> Hello,
> I wonder what analysis i have to use to evaluate which environmental
> variables most closely related to the grouping that I have.
> 
> I has 38 streams are grouped based on eight environmental variables, but I
> wonder how these variables relate to these groups.
> 
> Example.: PH, dissolved oxygen and altitude over which these variables
> relate to a group one, and two . . . 
> 
> Understood .....????
> 
> my email is vascomc at gamil.com
> 
> Thank you all.

If you clustered the 38 streams on the basis of these eight env
variables it seems a bit perverse to then ask how well these variables
then separate the groups.

For such a small sample, you might be best off computing summary
statistics for the 8 variables conditioned on the groups (mean of Var1
in groups A,B,C,... etc), accompanied by some graphical plotting of the
Variables (e.g box plots of VarX conditioned on group). Such an approach
would presume you aren't trying to predict group membership from the 8
variables.

HTH

G
-- 
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
 Dr. Gavin Simpson             [t] +44 (0)20 7679 0522
 ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
 Pearson Building,             [e] gavin.simpsonATNOSPAMucl.ac.uk
 Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
 UK. WC1E 6BT.                 [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%