Skip to content

A wrapper function for conditional inference forests

Source: R/party.R
cond_inference_surv_cforest.Rd

This function is a slightly different API for party::cforest() that has several important arguments as top-level arguments (as opposed to being specified in party::cforest_control()).

Usage

cond_inference_surv_cforest(
  formula,
  data,
  minsplit = 20L,
  maxdepth = 0L,
  teststat = "quad",
  testtype = "Univariate",
  mincriterion = 0,
  replace = FALSE,
  fraction = 0.632,
  mtry = 5L,
  ntree = 500L,
  ...
)

Arguments

formula

a symbolic description of the model to be fit. Note that symbols like : and - will not work and the tree will make use of all variables listed on the right-hand side of formula.

data

a data frame containing the variables in the model.

minsplit

the minimum sum of weights in a node in order to be considered for splitting.

maxdepth

maximum depth of the tree. The default maxdepth = 0 means that no restrictions are applied to tree sizes.

teststat

a character specifying the type of the test statistic to be applied.

testtype

a character specifying how to compute the distribution of the test statistic.

mincriterion

the value of the test statistic (for testtype == "Teststatistic"), or 1 - p-value (for other values of testtype) that must be exceeded in order to implement a split.

replace

a logical indicating whether sampling of observations is done with or without replacement.

fraction

fraction of number of observations to draw without replacement (only relevant if replace = FALSE).

mtry

number of input variables randomly sampled as candidates at each node for random forest like algorithms. The default mtry = 0 means that no random selection takes place.

ntree

number of trees to grow in a forest.

...

Other options to pass to party::cforest().

Value

An object of class RandomForest-class.

Details

Note that, although party::cforest_unbiased() is not directly used, the defaults for cond_inference_forest() mirror its default values. However, party::cforest_unbiased() does not allow several tuning parameters to be optimized (teststat, testtype, mincriterion, replace, and fraction). If you set pass a party::cforest_unbiased() object to cond_inference_forest() and modify those arguments, their values will be overwritten.

See also

cond_inference_surv_ctree library(survival) data(lung) cond_inference_surv_cforest( Surv(time, status) ~ age + sex, data = lung, mtry = 1, ntree = 10 )