simulate_inad: Simulate INAD antedependence series

In antedep: Antedependence Models for Longitudinal Data

Simulate INAD antedependence series

Description

Generate longitudinal count data from an INAD model using a thinning operator and an innovation distribution.

Usage

simulate_inad(
  n_subjects,
  n_time,
  order = 1L,
  thinning = c("binom", "pois", "nbinom"),
  innovation = c("pois", "bell", "nbinom"),
  alpha = NULL,
  theta = NULL,
  nb_inno_size = NULL,
  blocks = NULL,
  tau = 0,
  seed = NULL
)

Arguments

n_subjects	number of subjects
n_time	number of time points
order	antedependence order, 0, 1 or 2
thinning	thinning operator, one of "binom", "pois", "nbinom"
innovation	innovation distribution, one of "pois", "bell", "nbinom"
alpha	thinning parameter or vector or matrix; if NULL, defaults are used depending on the order
theta	innovation parameter or vector; if NULL, defaults are used depending on the innovation type. For Poisson and negative binomial innovations, theta is the innovation mean parameter. For Bell innovations, theta is the Bell rate parameter, with innovation mean theta * exp(theta).
nb_inno_size	size (dispersion) parameter for negative binomial innovations when innovation = "nbinom"; must be positive. If NULL, defaults to 1. Larger values correspond to less overdispersion (approaching Poisson as size -> Inf).
blocks	integer vector of length n_subjects indicating block membership for each subject; if NULL, no block effect is applied
tau	group effect vector indexed by block; tau[1] is forced to 0. If scalar x, it is expanded to c(0, x, ..., x) with length equal to the number of blocks
seed	optional random seed for reproducibility

Details

Time 1 observations are generated from the innovation distribution alone. For times 2 to n_time, counts are generated as thinning of previous counts plus independent innovations. When order = 0, all time points are generated from the innovation distribution and the thinning operator and alpha are ignored.

If blocks is provided, innovations include a block effect. For Poisson and negative binomial innovations, the innovation mean is theta[t] + tau[blocks[i]]. For Bell innovations, the innovation mean is theta[t] * exp(theta[t]) + tau[blocks[i]].

Value

integer matrix of counts with dimension n_subjects by n_time

Examples

y <- simulate_inad(
  n_subjects = 20,
  n_time = 6,
  order = 1,
  thinning = "binom",
  innovation = "pois",
  alpha = 0.3,
  theta = 2,
  seed = 42
)
dim(y)

antedep documentation built on April 25, 2026, 1:06 a.m.