predict.pollinated.ranger: Make predictions from a pollinated ranger random forest model

In saberpowers/causalLearning: Methods for heterogeneous treatment effect estimation

Description

Make predictions from a pollinated ranger random forest model

Usage

## S3 method for class 'pollinated.ranger'
predict(object, newx, predict.all = FALSE,
  na.treatment = c("omit", "replace", "NA"), ...)

Arguments

object	a fitted pollinated.ranger object
newx	matrix of new covariates for which predictions are desired
predict.all	logical: should predictions from all trees be returned? Otherwise the average across trees is returned
na.treatment	how to treat NA predictions from individual trees: 'omit' only uses trees for which the prediction is not NA. 'replace' replaces NA predictions with the overall mean response. 'NA' returns NA if any tree prediction is NA.
...	additional arguments passed on to predict.ranger

Value

a vector of predicted treatment effects corresponding to the rows of newx

saberpowers/causalLearning documentation built on May 30, 2019, 8:26 a.m.