predict.metafrontier: Predict Frontier Values from a Metafrontier Model

In metafrontier: Analysis of Metafrontier Models for Efficiency and Productivity

Predict Frontier Values from a Metafrontier Model

Description

Computes predicted frontier values at given input levels using either the metafrontier or a group-specific frontier.

Usage

## S3 method for class 'metafrontier'
predict(object, newdata = NULL, type = c("meta", "group"), ...)

Arguments

object	a "metafrontier" object.
newdata	an optional data frame of new inputs. If omitted, the training data predictions are returned.
type	character. "meta" (default) for metafrontier predictions, or "group" for group-specific frontier predictions (requires a group column in newdata).
...	additional arguments (currently unused).

Value

A numeric vector of predicted frontier values.

Examples

sim <- simulate_metafrontier(n_groups = 2, n_per_group = 50, seed = 42)
fit <- metafrontier(log_y ~ log_x1 + log_x2, data = sim$data,
                    group = "group", meta_type = "stochastic")
pred <- predict(fit)
# Out-of-sample prediction:
newdata <- data.frame(log_x1 = c(1, 2), log_x2 = c(1.5, 2.5))
predict(fit, newdata = newdata)

metafrontier documentation built on Aug. 19, 2026, 5:08 p.m.