Use the simulated ‘mstrata’ data (reminder: load these from within RMark) by the command
> require(RMark)
> data(mstrata)
Create several additional models to the ones already covered in class that describe combinations of stratum and time specific variation in survival, movement, and recapture probabilities.
Derive model-averaged estimates of the real parameters.
What is the Survival of the individuals from seasons 3 to 4, at the states A, B, and C?
What are the probabilities that an individual changes from state A to B, B to C, and A to C in the 3rd to 4th seasons? What are the probabilities that the individual stays in the states A, B, or C?
Which one was the best model?
Now, using that model build the reverse Multistrata model. Calculate for both models the probability of transitioning to a new state followed by the survival conditioned on being in that new state (i.e. Phi) . Hint: you can easily get the movement transitions for a given model by:
Psilist=get.real(mod,"Psi",vcv=TRUE)
Psivalues=Psilist$estimates
TransitionMatrix(Psivalues[Psivalues$time==1,])
and the survival parameters by
S.list=get.real(mod,"S",vcv=TRUE)