logLik.coxme {coxme} | R Documentation |
logLik is most commonly used for a model fitted by maximum likelihood, and some uses, e.g. by AIC. This method allows generic functions to easily extract the log-likelhood of a coxme model.
## S3 method for class 'coxme'
logLik(object, type = c("penalized", "integrated"), ...)
object |
a fitted coxme model |
type |
which of the two types of partial likelihood to extract |
... |
used by other methods |
The likelihood for a mixed effects Cox model can be viewed in two ways: the ordinarly partial likelihood, where the random effects act only as a penalty or constraint, or a partial likelihood where the random effect has been integrated out. Both are valid.
Returns an object of class logLik.