Compute confidence intervals for model parameters using profile likelihood.
Usage
# S3 method for class 'model_spec'
confint(object, parm, level = 0.95, y, par, ...)
Arguments
- object
A model_spec object
- parm
Parameter names (character vector). If missing, all parameters.
- level
Confidence level (default 0.95)
- y
Numeric vector of observed data
- par
Named numeric vector of parameter values (typically MLE)
- ...
Additional arguments (ignored)
Value
A matrix with columns for lower and upper bounds
Examples
spec <- model_spec(
loglik_fn = function(y, mu, sigma) sum(dnorm(y, mu, sigma, log = TRUE)),
par_names = c("mu", "sigma"),
par_bounds = list(sigma = c(1e-6, Inf))
)
set.seed(123)
y <- rnorm(100, mean = 5, sd = 2)
confint(spec, y = y, par = c(mu = 5, sigma = 2))