U-CARE Missing Data Dot (.) Notation

[moses502]

Does U-CARE handle the dot notation for missing data? I have a data set that spans 56 occassions (from 5 years) across 11 study sites, which were not all sampled in all years on all occassions.

[cooch]

Does U-CARE handle the dot notation for missing data? I have a data set that spans 56 occassions (from 5 years) across 11 study sites, which were not all sampled in all years on all occassions.

No - the 'dot' notation you're referring to is a MARK feature (currently undocumented, by and large). If you're considering U-CARE for GOF for MS models, you can estimate c-hat using MARK - although U-CARE gives more more 'diagnostic assistance' for sources of lack of fit.

[CHOQUET]

Hello,

the dot notation has no consistance at all and if confusing
should be removed by Evan in the Mark manual.
By definition, Capture-Recapture models handle missing data because
CR DATE include MISSING DATA as soon as capture rate is lower than 1.

In any case, can you give a good reference EVAN (article of course) for the method used in MARK to compute the chat ?.

As point out by EVAN, U-CARE gives more 'diagnostic assistance' for sources of lack of fit for multistate data and includes directional test
with more power.

Sincerely,

Rémi