Bayesian information criterion for fitted INAD models

Computes BIC using the fitted log likelihood and a parameter count that respects structural zeros and identifiability constraints.

Usage

bic_inad(fit, n_subjects = NULL)

Arguments

fit: A fitted model object returned by fit_inad.
n_subjects: Number of subjects, typically nrow(y). If NULL, inferred from fit$settings$n_subjects or legacy length(fit$settings$blocks) when available (with a warning).

Value

A numeric scalar BIC value.

Details

The BIC is computed as: $$BIC = -2 \times \ell + k \times \log(N)$$ where $\ell$ is the log-likelihood, $k$ is the number of free parameters, and $N$ is the number of subjects.

Examples

set.seed(1)
y <- simulate_inad(
  n_subjects = 40,
  n_time = 5,
  order = 1,
  thinning = "binom",
  innovation = "pois",
  alpha = 0.3,
  theta = 2
)
fit <- fit_inad(y, order = 1, thinning = "binom", innovation = "pois", max_iter = 20)
bic_inad(fit, n_subjects = nrow(y))
#> [1] 754.8662