Predict terminal-node outcome means from a greedy CI tree — predict_ci_tree_terminal

Assigns training observations to terminal nodes, computes a weighted mean outcome in each node, and returns those node means for new_data. This is a simple prediction rule for scoring ci_tree() models when the tree itself is primarily grown to reduce within-node inequality rather than prediction error.

Usage

predict_ci_tree_terminal_mean(
  fit,
  train_data,
  new_data = train_data,
  outcome_name,
  weights = NULL
)

predict_ctree_ci_terminal_mean(...)

Arguments

fit: A fitted ci_tree or other object inheriting from party.
train_data: Training data used to compute terminal-node means.
new_data: Data for which predictions should be returned. Defaults to train_data.
outcome_name: Name of the numeric or logical outcome column in train_data.
weights: Optional non-negative training weights.
...: Arguments passed to predict_ci_tree_terminal_mean().

Value

A numeric vector of predicted terminal-node means.

Examples

toy_data <- data.frame(
  rank = c(10, 20, 30, 40, 50, 60),
  outcome = c(1, 0, 1, 0, 1, 1),
  income = c(2, 4, 6, 8, 10, 12)
)
fit <- ci_tree(
  cbind(rank, outcome) ~ income,
  data = toy_data,
  rank_name = "rank",
  outcome_name = "outcome",
  control = ci_tree_control(minsplit = 1, minbucket = 1, maxdepth = 1)
)
predict_ci_tree_terminal_mean(fit, toy_data, outcome_name = "outcome")
#> [1] 0.6 0.6 0.6 0.6 0.6 1.0