BART
-package modelsGet predict draws from posterior of BART
-package models
predicted_draws_BART(
object,
newdata = NULL,
value = ".prediction",
...,
rng = stats::rnorm,
include_newdata = TRUE,
include_fitted = FALSE,
include_sigsqs = FALSE
)
A tidy data frame (tibble) with predicted values.
A BART
-package model.
Data frame to generate predictions from. If omitted, most model types will generate predictions from the data used to fit the model.
The name of the output column for predicted_draws
; default ".prediction"
.
Arguments to pass to predict
(e.g. BART:::predict.wbart
).
Random number generator function. Default is rnorm
for models with Gaussian errors.
Should the newdata be included in the tibble?
Should the posterior fitted values be included in the tibble?
Should the posterior sigma-squared draw be included?