Approximate Confidence Distribution Computing

Approximate confidence distribution computing (ACDC) offers a new take on the rapidly developing field of likelihood-free inference from within frequentist framework. The appeal this computational method for statistical hinges upon concept distribution, special type estimator which is defined with respect to repeated sampling principle. An ACDC provides validation in problems unknown or intractable likelihoods. main theoretical contribution work identification matching condition necessary validity method. In addition providing an example how modern understanding theory can be used connect Bayesian and inferential paradigms, we present case expand current scope so-called approximate include non-Bayesian by targeting rather than posterior. practical development data-driven approach drive both contexts. algorithm selection data-dependent proposal function, structure quite general adaptable many settings. We explore three numerical examples that verify arguments suggest instances outperform methods computationally.