survsrf_train {survcompare}

Fits randomForestSRC, with tuning by mtry, nodedepth, and nodesize. Underlying model is by Ishwaran et al(2008) https://www.randomforestsrc.org/articles/survival.html Ishwaran H, Kogalur UB, Blackstone EH, Lauer MS. Random survival forests. The Annals of Applied Statistics. 2008;2:841–60.

Description

Fits randomForestSRC, with tuning by mtry, nodedepth, and nodesize. Underlying model is by Ishwaran et al(2008) https://www.randomforestsrc.org/articles/survival.html Ishwaran H, Kogalur UB, Blackstone EH, Lauer MS. Random survival forests. The Annals of Applied Statistics. 2008;2:841–60.

Usage

survsrf_train(
  df_train,
  predict.factors,
  fixed_time = NaN,
  inner_cv = 3,
  randomseed = NULL,
  srf_tuning = list(),
  fast_version = TRUE,
  oob = TRUE,
  verbose = FALSE
)

Arguments

df_train

data, "time" and "event" should describe survival outcome

predict.factors

list of the column names to be used as predictors

fixed_time

time at which performance is maximized

inner_cv

k in k-fold CV for model tuning

randomseed

random seed

srf_tuning

list of mtry, nodedepth and nodesize, default is NULL

fast_version

TRUE/FALSE, TRUE by default

oob

TRUE/FALSE use out-of-bag predictions while tuning SRF instead of cross-validation, default is TRUE and is faster

verbose

TRUE/FALSE, FALSE by default

Value

output = list(beststats, allstats, model)