Predict an extremal random forest (ERF)

Makes prediction using a fitted erf object.

# S3 method for erf
predict(object, newdata = NULL, quantiles = c(0.95, 0.99), ...)

Arguments

object

Fitted erf object.

newdata

Numeric matrix or data.frame of test predictor values at which predictions are to be made. If NULL, predictions are made on the training data instead. For further information see grf::quantile_forest(). Default is NULL.

quantiles

Numeric vector of quantile levels at which estimates are to be made. Default is c(0.95, 0.99).

...

Dots. Currently ignored.

Value

Numeric matrix with predictions for each test point (rows) at the desired quantile levels (columns).