[R-sig-eco] AUC in MARS

[Rajendra Mohan Panda]

Dear All

Is there any way to calculate AUC/ROC for multiple adaptive regression
spline (MARS) for multiple response model ? I know it is possible  from the
published paper by Elith J. and Leathwick J. in Diversity and
Distributions (2007) 13, 265–275.

I have explored earth and mda package in this regard. But the R-codes
provided in EARTH notes by the same authors regarding roc or elsewhere are
not clear to my understandings.

It is not a problem for me to calculate AUC for individual models using
various modelling techniques (Biomod2 etc.). However, for multi-response
models, I am confused and need your help. Kindly advise how this can be
possible in a simplest way. Shall I use GIS? If yes please make it clear
how?

Your kind response is highly appreciated.


Best Regards
Rajendra M Panda
School of Water Resources
Indian Institute of Technology Kharagpur

	[[alternative HTML version deleted]]