[R-sig-eco] Quick Adonis question

[Gavin Simpson]

On Thu, 2011-04-07 at 06:21 +0000, Nancy Shackelford wrote:
> Hi everyone,
> 
> I have a huge number of variables and am trying to find the best model for my
> data including second order interactions.  I just wanted to know if there's a
> way to have adonis output the AIC of a model rather than the r^2 values.  Or I
> suppose another solution to my problem would be a 'step' function for adonis
> such as that used with glm.  Does that exist?

Can you direct us to a definition of AIC for the sorts of models fitted
via adonis()?

AIC requires a likelihood and I don't think I've come across one of
those for an ordination recently. Someone did do this for CCA (IIRC) and
it has been implemented in vegan() but that is as far as things go.

By the way, it does sound somewhat like you are entering data-dredging
territory with what you describe. If I were doing this, I'd take a step
back and think about plausible models and fit those rather than trying
to step through all my variables to see what crops up.

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
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%