Timeline for Camera trapping and the Poisson distribution
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 4, 2019 at 22:45 | comment | added | Ben | Deviation from the Poisson is likely to come from two sources: (1) the fact that different animals appear with different rates, so you are aggregating over different rates; and (2) the fact that different days have different seasons/temps, etc., and so also give different rates. It is therefore likely that the distribution will be some kind of mixture of Poisson distributions, such as the negative binomial. | |
| Jun 4, 2019 at 12:53 | comment | added | David | @Ben But the data comes from per-day aggregation on a similar time of the year, so I see no reason to believe that there are "special days" | |
| Jun 4, 2019 at 12:14 | comment | added | Ben | @David: The animals positions are almost certainly neither uniform, nor independent of one another. Hence, I doubt there is any good theoretical reason to believe in a Poisson distribution. | |
| Jun 4, 2019 at 11:00 | comment | added | Scortchi♦ | (+1) I'd note, though, that studies are often contrived to eliminate - well, reduce to negligible levels - heterogeneity: its presence hints at factors that can be controlled or measured & modelled. And - I think this is what @David is saying in the comment above - the whole point might be to test the adequacy of the model. | |
| Jun 4, 2019 at 10:13 | comment | added | David | Indeed, there is a strong theoretical reason to model this as a Poisson process, since that would be the distribution we'd have if the animals' poistions were random | |
| Jun 4, 2019 at 0:48 | history | answered | Ben | CC BY-SA 4.0 |