predict.beezdemand_hurdle: Predict Method for Hurdle Demand Models

In beezdemand: Behavioral Economic Easy Demand

Predict Method for Hurdle Demand Models

Description

Returns predictions from a fitted hurdle demand model.

Usage

## S3 method for class 'beezdemand_hurdle'
predict(
  object,
  newdata = NULL,
  type = c("response", "link", "parameters", "probability", "demand"),
  prices = NULL,
  se.fit = FALSE,
  interval = c("none", "confidence"),
  level = 0.95,
  ...
)

Arguments

object	An object of class beezdemand_hurdle.
newdata	Optional data frame containing a price column matching the fitted object's x_var. If newdata includes the id column, subject-specific predictions are returned; otherwise population predictions are returned. If newdata is NULL, returns predictions for all subjects across a price grid.
type	One of: "response" Predicted consumption (part II) "link" Predicted log-consumption (linear predictor of part II) "probability" Predicted probability of zero consumption (part I) "demand" Predicted expected consumption = (1 - P0) * response "parameters" Subject-specific parameters (no .fitted column)
prices	Optional numeric vector of prices used only when newdata = NULL.
se.fit	Logical; if TRUE, includes a .se.fit column (delta-method via sdreport when available).
interval	One of "none" (default) or "confidence".
level	Confidence level when interval = "confidence".
...	Unused.

Value

For type = "parameters", a tibble of subject-level parameters. Otherwise, a tibble containing the newdata columns plus .fitted and helper columns predicted_log_consumption, predicted_consumption, prob_zero, and expected_consumption. When requested, .se.fit and .lower/.upper are included.

Examples

data(apt)
fit <- fit_demand_hurdle(apt, y_var = "y", x_var = "x", id_var = "id")

# Get subject-specific parameters
pars <- predict(fit, type = "parameters")

# Predict demand at specific prices
demand <- predict(fit, type = "demand", prices = c(0, 0.5, 1, 2, 5))

beezdemand documentation built on March 3, 2026, 9:07 a.m.