aggregation.pmpec  R Documentation 
Interface to pmpec
, for conforming to the structure required by the argument aggregation.fun
in peperr
call. Evaluates the prediction error curve, i.e. the Brier score tracked over time, for a fitted survival model.
aggregation.pmpec(full.data, response, x, model, cplx=NULL, times = NULL,
type=c("apparent", "noinf"), fullsample.attr = NULL, ...)
full.data 
data frame with full data set. 
response 
Either a survival object (with 
x 

model 
survival model as returned by 
cplx 
numeric, number of boosting steps or list, containing number of boosting steps in argument 
times 
vector of evaluation time points. If given, used as well as in calculation of full apparent and noinformation error as in resampling procedure. Not used if 
type 
character. 
fullsample.attr 
vector of evaluation time points, passed in resampling procedure. Either userdefined, if 
... 
additional arguments passed to 
If no evaluation time points are passed, they are generated using all uncensored time points if their number is smaller than 100, or 100 time points up to the 95% quantile of the uncensored time points are taken.
pmpec
requires a predictProb
method for the class of the fitted model, i.e. for a model of class class
predictProb.class
.
A matrix with one row. Each column represents the estimated prediction error of the fit at the time points.
peperr
, predictProb
, pmpec
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.