Dear Miltinho,
It depends on the species (rare or widespread), and also on how well its
distribution is known.
You have to make some assumptions about where the species might not be
present and generate the pseudo-absences in these areas. However, other
people just simply create pseudo-absence data from the overall study area
without confirmed presence of the species.
I can not help you with the appropriate reference to your question, because
to my knowledge there is not a specific paper addressing this issue. But as
I said, check in the ISI Web of Science and some paper discuss similar
topics.
Cheers,
Núria
2008/9/20 milton ruser <milton.ruser@gmail.com>
> Dear Núria,
> Thank you very much for your reply.
> I will try follow your suggestions and generate a pseudoabsence list.
> By the way, have you some suggestion of reading on this regard?
> Taking in account that I don´t have know-absence, how technics can I use
> do sort random pseudoabsence points?
>
> Best wishes,
>
> miltinho
> brazil
>
> On Sat, Sep 20, 2008 at 2:18 PM, Núria Roura Pascual <
> nrouralists@gmail.com> wrote:
>
>> Dear Miltinho,
>>
>> Since you don´t have absence, I think that you cannot calculate the
AUC.
>> However, you can create pseudoabsences (by selecting areas from where
you
>> know that the species is not present) at random and use them as
surrogates
>> of absences data. This process has its drawbacks, which have been well
>> described in the scientific literature.
>>
>> Hope it helps. Cheers,
>>
>> Núria
>>
>> 2008/9/20 milton ruser <milton.ruser@gmail.com>
>>
>> > Dear all,
>> >
>> >
>> >
>> > I have a probability of presence of distribution of a species of
>> interest
>> >
>> > (varying from 0 to 1 in continuous form) and I have a set of
points
>> >
>> > where I know that species really occurs. But I don´t have points
of
>> > absence.
>> >
>> >
>> >
>> > So, for each true presence I know the estimated presence.
>> >
>> > I would like to know how can I compute AUC, taking account these
>> >
>> > Available data.
>> >
>> >
>> >
>> > Best wishes,
>> >
>> >
>> >
>> > Miltinho autronauta
>> >
>> > brazil
>> >
>> > [[alternative HTML version deleted]]
>> >
>> >
>> > ______________________________________________
>> > R-help@r-project.org mailing list
>> > https://stat.ethz.ch/mailman/listinfo/r-help
>> > PLEASE do read the posting guide
>> > http://www.R-project.org/posting-guide.html
>> > and provide commented, minimal, self-contained, reproducible code.
>> >
>> >
>>
>> [[alternative HTML version deleted]]
>>
>>
>> ______________________________________________
>> R-help@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>>
>
[[alternative HTML version deleted]]