Contents
- Index
Barker Model Parameterizations
Two alternative parameterizations are discussed by Barker (1997). The relationship with the MARK parameterizations is shown in the following tables.
1. Parameterization in terms of theta(i), f(i), nu(i)
Parameter Description Relationship with Mark
Parameters
theta(i) Pr(not resighted in i,i+1 | alive at i and i+1) theta(i) = 1-R(i)
f(i) Pr(resighted (alive or dead) in i,i+1 | alive at i) f(i) = S(i)R(i) +
(1-S(i))*{r(i) + (1-r(i))R'(i)}
nu(i) Pr(resighted alive | resighted in i,i+1) 1-(1-S(i))r(i)/f(i)
This parameterization was motivated by a trout study where "resightings" were captures by anglers between the live-recapture samples. Fish alive at time i were captured by anglers with probability f(i), then either released with probability nu(i) or killed with probability (1-nu(i)).
2. Parameterization in terms of theta(i), rho(i), r(i)
Parameter Description Relationship with Mark
Parameters
theta(i) Pr(not resighted in i,i+1 | alive at i and i+1) theta(i) = 1-R(i)
rho(i) Pr(resighted alive in i,i+1 | alive at i rho(i) = S(i)R(i) + (1- r(i))R'(i)
r(i) Pr(tag is reported | found dead in i,i+1) r(i) = r(i)
Also, in terms of the f(i) and nu(i) parameterization:
rho(i) = f(i)nu(i) ,
and
r(i) = f(i)(1-nu(i))/(1-S(i)) .
The rho(i), r(i) parameterization is motivated by studies where animals are either resighted alive or found dead.