I have birds translocated from site A (original, capture site) to site B (new, release site) and I want to analyse homing behaviour. Translocation of individuals was performed continuously (several weeks), thus a bird released at the beginning of the experiment had theoretically highest chance to be recaptured at the original site comparing to a bird which was released before the end of the experiment. My data represents date of capture(=date of translocation and release) and date of recapture. Is there correct method how to estimate probability of recapture in a population (how to include such time effect)? Would it be possible to simulate homing somehow (how fast they come home)? Thank you for any guidance in advance!
1
-
$\begingroup$ +1 "guidance" : the process of controlling the flight of something (such as a [pigeon]) (from m-w.com) $\endgroup$– Jack RyanCommented Mar 27, 2014 at 14:07
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
1
I think that what you have is a basic Survival Analysis problem. (According to Wikipedia, This topic is called reliability theory or reliability analysis in engineering, and duration analysis or duration modeling in economics or event history analysis in sociology).
Your response variable seems to be time to event with censoring to the right, that is, there are some homecomings you don't register because your experiment finishes before (or maybe because they never happen). What you can obtain with this analysis is the probability of being recaptured given time from release and any other covariates.
You can try with library survival in R, if that's your software of choice. Many other softwares have equivalent tools to work with this kind of data.
-
$\begingroup$ thank you! It seems that this could be applicable for my problem. I will play around a bit. $\endgroup$ Commented Mar 28, 2014 at 12:30