RISCA (version 1.0.1)

summary.sl.time: Summaries of a Super Learner

Description

Return goodness-of-fit indicators of a Super Learner obtained by the function sl.time.

Usage

# S3 method for sl.time
summary(object, ..., method, pro.time, newdata, times, failures)

Arguments

object

An object returned by the function sl.time.

...

Additional arguments affecting the plot.

method

A character string with the name of the algorithm included in the SL for wich the calibration plot is performed. The defaut is "sl" for the Super Learner.

pro.time

The prognostic time up to which the time-dependent indicators are estimated.

newdata

An optional data frame containing the new sample for validation with covariate values, folow-up times, and event status. The default value isNULL, the calibration plot is performed from the same subjects of the training sample.

times

The name of the variable related the numeric vector with the follow-up times in newdata (optional argument only necessary when newdata is not NULL).

failures

The name of the variable related the numeric vector with the event indicators (0=right censored, 1=event) in newdata (optional argument only necessary when newdata is not NULL).

Author

Yohann Foucher <[e-mail]>

Camille Sabathe <[e-mail]>

Details

The following metrics are returned: "brier" for the Brier score at the prognostic time pro.time, "loglik" for the Log-likelihood, "ibs" for the Integrated Brier score up to the last observed time of event, "ibll" for the Integrated Binomial Log-likelihood up to the last observed time of event, "bll" for the binomial Log-likelihood, "ribs" for the restricted Integrated Brier score up to the prognostic time pro.time, "ribll" for the restricted Integrated Binomial Log-likelihood Log-likelihood up to the last observed time of event, "bll" for the binomial Log-likelihood, "auc" for the area under the time-dependent ROC curve up to the prognostic time pro.time.

Examples

data(dataDIVAT2)

dataDIVAT2$train <- 1*rbinom(n=dim(dataDIVAT2)[1], size = 1, prob=1/3)


# The training of the super learner with 2 algorithms
sl<-sl.time(method=c("aft.gamma", "ph.gompertz"),  metric="ibs",
  data=dataDIVAT2[dataDIVAT2$train==1,],  times="times", failures="failures", pro.time = 12,
  cov.quanti=c("age"),  cov.quali=c("hla", "retransplant", "ecd"),  cv=3)

# The prognostic capacities from the validation sample
summary(sl, method="sl", pro.time=12,
     newdata=dataDIVAT2[dataDIVAT2$train==0,],
     times="times", failures="failures")