Open tommyvcc opened 1 year ago
the akaike information criterion is not well defined with the cox partial likelihood.
AIC is defined in terms of the full likelihood, the partial likelihood is different from that
Thank you so much for your support and prompt response.
Is there anyway to calculate the Akiki information criterion or the full likelihood for the deep cox mixtures model?
Dear all
After running the deep cox mixtures, how can I calculate or know the log partial likelihood? Is there a code for that?
The reason I am asking for it is to calculate the Akaiki Information criterion and compare it with other models
Would you kindly be able to help me with that point?