Is there a difference between randomtestpoints argument and subsetting your own test/train data?

[Lauren Yee]

Hi all,

I've been using the argument randomtestpoints of 30% for my maxent models, however, I am wondering if there is a difference between doing that and doing what is in the dismo documentation:

# witholding a 20% sample for testing
fold <- kfold(occ, k=5)
occtest <- occ[fold == 1, ]
occtrain <- occ[fold != 1, ]
# fit model, biome is a categorical variable
me <- maxent(predictors, occtrain, factors='biome')

Is there any difference or benefits to using one method over the other?

Thanks!

[Jamie M. Kass]

If you use dismo's kfold() with k=5 to randomly partition your occurrence points like you did in your example, and specify randomtestpoints as 20% with maxent.jar, these should be equivalent.

-Jamie