GOF test for known fate

[Darlene Kilpatrick]

Hopefully I will have a better response to this question than my last one.

It is very simple. I am running known fate for some of my data. I am trying to test for goodness of fit. I read the manual where it suggests running median c hat. It does not specify if it can be run for all types of data analysis. It mentions CJS. I have tried to run median c hat for 3 known fate models and it does not work- at least for my data. On the window that says the percent completed it remains blank no matter how long I run it.

Do you not test GOF for known fate models? Is there a better method for known fate or is something screwy with my data?

I'ld appreicate some help. Thanks!
Darlene Kilpatrick

[cooch]

Do you not test GOF for known fate models? Is there a better method for known fate or is something screwy with my data?

No - for a simple reason. For known-fate data (and nest survival data, which is a form of a known fate analysis), the saturated model is a useful model. Because the saturated model fits the data perfectly, there is no GOF test. To obtain a GOF test, we would have to assume a reasonable biological model that is reduced from the saturated model. This selection can be pretty arbitrary, needless to say.

I suppose its time to put an extended discussion of saturated models and GOF in the GOF chapter...