Using simulation to handle implicit likelihoods in a Bayesian analysis
Michael S. Hamada, Timothy Graves, Nicolas Hengartner, David Higdon, Aparna V. Huzurbazar, Earl Lawrence, Crystal D. Linkletter, C. Shane Reese, D. W. Scott, Randy R. Sitter, Richard L. Warr, Brian J. Williams · Quality Engineering · 2021
KEY POINTSThis article presents a Bayesian inferential method where the likelihood for a model is unknown, i.e., an implicit likelihood, but where data can easily be simulated from the data model. We use simulated data to estimate the implicit likelihood in a Bayesian analysis employing a Markov chain Monte Carlo algorithm. Two examples are presented.