simulate_mnl_data: Simulate multinomial logit data

In choicer: Discrete Choice Models for Economic Applications

Simulate multinomial logit data

Description

Generates synthetic choice data with i.i.d. Gumbel errors, optionally with varying choice-set sizes and an outside option (alt = 0). Choices are determined by argmax of utility; covariates are drawn as Uniform(-1, 1).

Usage

simulate_mnl_data(
  N = 5000,
  J = 5,
  beta = c(0.8, -0.6),
  delta = NULL,
  seed = 123,
  outside_option = TRUE,
  vary_choice_set = TRUE
)

Arguments

N	Number of choice situations.
J	Number of inside alternatives.
beta	Fixed coefficients for ⁠x1..x{K_x}⁠ (length K_x = length(beta)).
delta	Alternative-specific constants for inside alternatives (length J). Defaults to an alternating pattern of c(0.5, -0.5).
seed	Random seed. Pass NULL to skip set.seed() (useful inside monte_carlo() where the caller manages RNG).
outside_option	Logical; if TRUE (default) an outside option with alt = 0 and zero covariates is added to every choice set.
vary_choice_set	Logical; if TRUE (default) choice set size is sampled uniformly from 2:J; if FALSE every individual faces all J inside alternatives.

Value

A choicer_sim object.

Examples

sim <- simulate_mnl_data(N = 1000, J = 5, seed = 123)
print(sim)

choicer documentation built on Sept. 5, 2026, 1:07 a.m.