Median c-hat procedure with dots (missing sessions)

[rdanner]

Dear Community,

I couldn't find anything on the forum or in the Book about this...

I'm modeling phi and p with CJS models using an encounter history data set where some sessions were not completed, so are represented as dots.

When I start the Median c-hat procedure, MARK gives the following warning: “Warning – Dots in the encounter histories are not properly handled with the bootstrap or median c-hat simulations.” If you go ahead with the bootstrapping, however, it does give a c-hat value. Does anyone know if it would be better to use this median c-hat value despite dots being mishandled, or to skip the median c-hat procedure altogether?

Not sure if it matters, but I built the models with RMark and exported them to MARK for the median c-hat analysis.

Thanks,
Ray

[cooch]

Dear Community,

I couldn't find anything on the forum or in the Book about this...

I'm modeling phi and p with CJS models using an encounter history data set where some sessions were not completed, so are represented as dots.

When I start the Median c-hat procedure, MARK gives the following warning: “Warning – Dots in the encounter histories are not properly handled with the bootstrap or median c-hat simulations.” If you go ahead with the bootstrapping, however, it does give a c-hat value. Does anyone know if it would be better to use this median c-hat value despite dots being mishandled, or to skip the median c-hat procedure altogether?

Not sure if it matters, but I built the models with RMark and exported them to MARK for the median c-hat analysis.

Thanks,
Ray

It means what it says -- for missing values, there is no entirely robust way of estimating c-hat. Period. If your estimated c-hat is >1, then at worst, using it will make your analysis 'more conservative'. Choice is yours.

[rdanner]

Thanks!

[Fish_Boy]

I may be wrong, but could you not simply remove the dots and set the time interval to the appropriate length? i.e., skip the occasion with incomplete information. It has been a while since I RTFM.