Degrees of freedom UCARE

[Gingko]

Hello,
I'm trying to make Goodness-of-fit test on a dataset of multistate capture-recapture data. I found that there was a difference in the number of degrees of freedom calculated by U-CARE 2.0.2 compare to U-CARE 2.2.5. I analyse the same dataset and when performing the test for a fit to the JMV model, I have 28 degrees of freedom with U-CARE 2.0.2 and 45 degrees of freedom with U-CARE 2.2.5 (for a given group of individuals)
Are the tests performed in the same way in both versions of the software?
In fact, what I want to do is test for a fit to the CAS model. I got back to the version 2.0.2 because I couldn't find how to perform this test in the version 2.2.5. Is there a way to do it?
Thanks for your help,
Gingko

[CHOQUET]

Hello,
there is a big difference in the strategy for pooling the tables between the two versions of u-care. This is the main reason why the numbers of degrees of freedom are differents between the two versions.
Please use the lastest version of u-care which is more reliable.
We removed in the last version of u-care the LRT test between time dependent CAS and JMV models because most of the time and due to a lack of data, we are not able to fit these models.
Sincerely,
Rémi