[R-sig-eco] co-correspondence package

[Gavin Simpson]

On Mon, 2010-09-27 at 03:02 -0700, wout opdekamp wrote:
> Dear list,
> 
> I'm using the cocorresp-package as well and was wondering if any R analog
> for the predictive correspondence analysis and cross valuation is already
> included in the package?
> 
> Kind regards,
> 
> Wout

Dear Wout,

Yes, if you meant "predictive CO-correspondence analysis and cross
valuation", this has been there since the first release IIRC. In fact,
predictive Co-CA is the default in the main function coca().

coca() is the main function at it can do predictive CoCA by setting
'method = "predictive"' (which is the default so you don't actually need
to add this). The argument 'reg.method' controls which regression method
is used (SIMPLS or an eigen analysis version of SIMPLS. SIMPLS is faster
than eigen).

The cross-validation is done using function crossval().

I direct you to the relevant help pages ?coca and ?crossval for further
details.

This package has languished a bit since my initial bout of
enthusiasm/coding. If you have comments for improving the package,
please let me know.

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