search

wpeterman / ResistanceGA

Optimize resistance surfaces using Genetic Algorithms
37 stars 15 forks source link

Drastically lower avg. R2m with Resist.boot function, then with MS_optim/SS_optim #37

Closed hannehaug closed 1 year ago

hannehaug commented 1 year ago

Hi, I'm not sure if this is an issue with the software or with my data. The issue is that when running the MS_optim or SS_optim function I will get R2m values around 0.60 to 0.70. However, when running the Resist.boot function I get avg. R2m around 0.01 for all models. The dataset consist of 18 populations, so should be acceptable I think.

I have read the tutorial, which explains that the Resist.boot function may reveal if the observed pattern is driven by one or two samples. I wonder if that is what I am observing, but I'm not sure because the difference in R2m seems so extreme.

Any input on this?

hannehaug commented 1 year ago

Hi again, it was neither the data nor the software, but me not being able to spell correctly. I'm very sorry if I've caused any inconvenience. Feel free to delete this thread, if you want.