Conditional distribution from an MLE.

Computes the conditional distribution of a subset of parameters given observed values for other parameters. Uses the closed-form Schur complement for the multivariate normal. Returns a dist object (not an mle), since conditioning loses the MLE context.

Usage

# S3 method for class 'mle'
conditional(x, P = NULL, ..., given_indices = NULL, given_values = NULL)

Arguments

x

An mle object with at least 2 parameters.

P

Optional predicate function for Monte Carlo fallback.

...

Additional arguments forwarded to P.

given_indices

Integer vector of conditioned parameter indices.

given_values

Numeric vector of observed values.

Value

A normal, mvn, or empirical_dist object.