choice_probabilities: Define choice probabilities
In choicedata: Working with Choice Data

choice_probabilities

R Documentation

Define choice probabilities

Description

The choice_probabilities object defines the choice probabilities.

compute_choice_probabilities() calculates the choice probabilities based on the choice parameters and the choice data.

Usage

choice_probabilities(
  data_frame,
  choice_only = TRUE,
  column_decider = "deciderID",
  column_occasion = NULL,
  cross_section = FALSE,
  column_probabilities = if (choice_only) "choice_probability"
)

compute_choice_probabilities(
  choice_parameters,
  choice_data,
  choice_effects,
  choice_only = FALSE,
  input_checks = TRUE,
  ...
)

Arguments

`data_frame`	[`data.frame`] Contains the choice probabilities.
`choice_only`	[`logical(1)`] Only the probabilities for the chosen alternatives?
`column_decider`	[`character(1)`] The name of the identifier column for deciders.
`column_occasion`	[`character(1)` \| `NULL`] The name of the identifier column for choice occasions (panel data). Can be `NULL` for the cross-sectional case.
`cross_section`	[`logical(1)`] Treat choice data as cross-sectional?
`column_probabilities`	[`character()`] The column name of the `data_frame` with the choice probabilities for all choice alternatives. If `choice_only = TRUE`, it is the name of a single column that contains the probabilities for the chosen alternatives.
`choice_parameters`	[`choice_parameters` \| `list`] Either a `choice_parameters` object or a list in optimization space as returned by `switch_parameter_space`.
`choice_data`	[`choice_data`] A `choice_data` object providing responses and covariates.
`choice_effects`	[`choice_effects`] A `choice_effects` object defining the specification.
`input_checks`	[`logical(1)`] Should additional internal input checks be performed before computing the probabilities?
`...`	Passed to the underlying probability computation routine.

Value

A choice_probabilities S3 object (a data frame) that stores additional metadata in attributes such as column_probabilities, choice_only, and the identifier columns. These attributes are used by downstream helpers to reconstruct the original structure.

Examples

data(train_choice)
choice_effects <- choice_effects(
  choice_formula = choice_formula(
    formula = choice ~ price | time,
    error_term = "probit"
  ),
  choice_alternatives = choice_alternatives(
    J = 2, alternatives = c("A", "B")
  )
)
choice_parameters <- generate_choice_parameters(choice_effects)
choice_data <- choice_data(
  data_frame = train_choice,
  format = "wide",
  column_choice = "choice",
  column_decider = "deciderID",
  column_occasion = "occasionID"
)
compute_choice_probabilities(
  choice_parameters = choice_parameters,
  choice_data = choice_data,
  choice_effects = choice_effects,
  choice_only = TRUE
)

choicedata documentation built on Nov. 5, 2025, 5:46 p.m.