When you fit the logistic regression, you have to retain a random set of presence /absence points for validating the model (i.e. a dataset "Test", including a column "PA" representing your presence absence TEST points and other columns for the covariates of interest ).

Here an example of a code that should do the job. I hope it works (you have of course to change something depending on your model/dataset) 

## fit your model

M1

Hi Ram,

Have a look at the sensitivity package for R. There are detailed descriptions of each function you may want to check out, and specific tests you may want to conduct. It seems like a good place to start.

If you need any further assistance, feel free to ask.

Kind regards,

https://cran.r-project.org/web/packages/sensitivity/sensitivity.pdf