Marginal Likelihood Estimation via Arrogance Sampling

This paper describes a method for estimating the marginal likelihood or Bayes fac-tors of Bayesian models using non-parametric importance sampling (“arrogance sam-pling”). This method can also be used to compute the normalizing constant of probabil-ity distributions. Because the required inputs are samples from the distribution to be normalized and the scaled density at those samples, this method may be a convenient replacement for the harmonic mean estimator. The method has been implemented in the open source R package margLikArrogance.