Predictive partial log-likelihood for a CoxBoost model fit

Extracts the predictive partial log-likelihood from a CoxBoost model fit.

Usage

# S3 method for class 'CoxBoost'
PLL(object, newdata, newtime, newstatus, complexity, ...)

object: fitted model of class CoxBoost.
newdata: n_new*p matrix of covariates.
newtime: n_new-vector of censored survival times.
newstatus: n_new-vector of survival status, coded with 0 and 1.
complexity: complexity, either one value, which is the number of boosting steps, or a list containing stepno.
...: additional arguments passed to predict.CoxBoost.

Used by function peperr, if function fit.CoxBoost is used for model fit.

Since CoxBoost is only suggested by peperr, install it before calling this function.

Numeric predictive partial log-likelihood returned by predict.CoxBoost.