ggRandomForests (version 1.1.2)

gg_survival.ggRandomForests: Nonparametric survival estimates.

Description

Nonparametric survival estimates.

Usage

gg_survival.ggRandomForests(interval, censor, strat = NULL, data,
  type = c("kaplan", "nelson"), ...)

Arguments

interval

name of the interval variable in the training dataset.

censor

name of the censoring variable in the training dataset.

strat

stratifying variable in the training dataset, defaults to NULL

data

name of the training data.frame

type

one of ("kaplan","nelson"), defaults to kaplan-meier

...

extra arguments passed to kaplan or nelson functions.

Value

• A gg_survival object created using the non-parametric kaplan-meier or nelson-aalon estimators.

Details

gg_survival is a wrapper function for generating nonparametric survival estimates using either nelson-aalen or kaplan-meier estimates.

See Also

kaplan nelson plot.gg_survival

Examples

data(pbc, package="randomForestSRC")
pbc$time <- pbc$days/364.25

# This is the same as kaplan
gg_dta <- gg_survival(interval="time", censor="status",
                     data=pbc)

plot(gg_dta, error="none")
plot(gg_dta)

# Stratified on treatment variable.
gg_dta <- gg_survival(interval="time", censor="status",
                     data=pbc, strat="treatment")

plot(gg_dta, error="none")
plot(gg_dta)

# ...with smaller confidence limits.
gg_dta <- gg_survival(interval="time", censor="status",
                     data=pbc, strat="treatment", conf.int=.68)

plot(gg_dta, error="lines")