Akaike Information Criterion for a SIR Model

Computes AIC = -2 * log-likelihood + k * (number of parameters). Use this to compare SIR models with different specifications (e.g., different numbers of influence covariates).

Usage

# S3 method for class 'sir'
AIC(object, ..., k = 2)

Arguments

object: A fitted sir object from sir.
...: Additional arguments for comparison with other models.
k: Numeric penalty per parameter (default 2 for standard AIC).

Value

Numeric AIC value. Lower is better.