mobforest.control: Control parameters for random forest
In RTIInternational/mobForest: Model Based Random Forest Analysis

Description Usage Arguments Details Value References Examples

Various parameters that control the forest growing.

mobforest.control(ntree = 300, mtry = 0, replace = FALSE,
  fraction = 0.632, alpha = 1, bonferroni = FALSE, minsplit = 20,
  trim = 0.1, objfun = deviance, breakties = FALSE, parm = NULL,
  verbose = FALSE)

`ntree`	Number of trees to be constructed in forest (default = 300).
`mtry`	Number of input variables randomly sampled as candidates at each node.
`replace`	logical. replace = TRUE (default) performs bootstrapping. replace = FALSE performs sampling without replacement.
`fraction`	Number of observations to draw without replacement (only relevant if replace = FALSE).
`alpha`	A node is considered for splitting if the p value for any partitioning variable in that node falls below alpha (default 0.05). Please see mob_control().
`bonferroni`	logical. Should p values be Bonferroni corrected? (default TRUE). Please see mob_control().
`minsplit`	An integer. The minimum number of observations in a node (default 20). Please see mob_control().
`trim`	A numeric, as defined in mob_control().
`objfun`	A function, as defined in mob_control().
`breakties`	A logical, as defined in mob_control().
`parm`	A numeric or vector, as defined in mob_control().
`verbose`	A logical, as defined in mob_control().

This function is used to set up forest controls. The mob_control (from party 'package') object is used to set up control parameters for single tree model. For most parameters, please see: mob_control()

An object of class mobforest.control.

Achim Zeileis, Torsten Hothorn, and Kurt Hornik (2008). Model-Based Recursive Partitioning. Journal of Computational and Graphical Statistics, 17(2), 492-514.

1
2
3

# create forest controls before starting random forest analysis
mobforest_control = mobforest.control(ntree = 400, mtry = 4, replace = TRUE,
    minsplit = 200)