cross validation - Pearson's p value approach to test model's significance (pValueCompute.exe)

Question

I have two questions regarding the calculation of p value and success rate, method stated in Pearson et al., 2007.

1. Does this testing method work only when cross-validation is used in maxent setting? can it work if I used bootstrap?

If yes then, I noticed in all of the examples I came cross that the values of the first column (obtained from Minimum training presence test omission) are always 0 or 1 while in my file I have some with decimal value .. so could I say that since the 0 is converted to 1, so the 0.2 in my file will be converted to o.8 ?

Also I’m using the MTP (minimum training presence) threshold. It was not clear for me if I should apply the MTP threshold in maxent settings before running the model or keep it with no threshold?

Answer 1

等待大神答复