RDHHET Model Output Query

[K.Haigh]

Hi,

I am slowly teaching myself how to use RMark and I have just run the following RDHHet no movement model:

Code: Select all

#Parameter specifications for model formulas
S.dot=list(formula=~1)
p.session.mixture=list(formula=~session+mixture,share=TRUE)
pi.dot=list(formula=~1)
GammaDoublePrime.zeroTE=list(formula=~1,fixed=0)
GammaPrime.zeroTE=list(formula=~1,fixed=0)

#No movement model w/ constant survival, p=c varies by season (session)
No.mov.mod.5=mark(data=rd.process,
                  ddl=rd.ddl,
                  model.parameters=list(
                    S=S.dot,
                    p=p.session.mixture,
                    pi=pi.dot,
                    GammaDoublePrime=GammaDoublePrime.zeroTE,
                    GammaPrime=GammaPrime.zeroTE))

In the model output for Real Parameters, parameter p is listed as:

p gFemale s1 t0 m1
p gFemale s1 t0 m2
p gFemale s2 t0 m1
p gFemale s2 t0 m2
p gFemale s3 t0 m1
p gFemale s3 t0 m2
etc etc.....

Firstly, am I correct in assuming the 's' stands for 'session' (i.e., primary period), 't' stands for time (i.e., secondary period which is 0 as I have not specified 'time' in my model)? I am unsure what the m's stand for, are these the mixtures?

Secondly, I have three groups in my dataset (female, male, and unknown sex). Is this output showing me p estimates for the females only, or is it showing p estimates for all individuals in the dataset (as I have not specified group in my parameter specifications in the model) and simply labelling g as female?

[jlaake]

Those labels are not useful any longer and I probably should remove them. Use summary(model) or summary (model)$reals or you can add se=TRUE. You will get all estimates.

Some history for you. Originally RMark was designed with all different PIMS and all real parameters were maintained in the model output. This became untenable with large models so I created the simplification process which simplifies to the unique real parameters. This made the parameter labels useless because the labels may represent more than one of the original parameters. As you have seen female is used as label for all sexes since sex is not in the model.

m represents mixture. If you get reals from summary all reals can be retrieved but are not necessarily unique.

Hope this helps.

Jeff