Inference for Trans-dimensional Bayesian Models with Diffusive Nested Sampling
Abstract
Many inference problems involve inferring the number of components in some region, along with their properties , from a dataset . A common statistical example is finite mixture modelling. In the Bayesian framework, these problems are typically solved using one of the following two methods: i) by executing a Monte Carlo algorithm (such as Nested Sampling) once for each possible value of , and calculating the marginal likelihood or evidence as a function of ; or ii) by doing a single run that allows the model dimension to change (such as Markov Chain Monte Carlo with birth/death moves), and obtaining the posterior for directly. In this paper we present a general approach to this problem that uses trans-dimensional MCMC embedded within a Nested Sampling algorithm, allowing us to explore the posterior distribution and calculate the marginal likelihood (summed over ) even if the problem contains a phase transition or other difficult features such as multimodality. We present two example problems, finding sinusoidal signals in noisy data, and finding and measuring galaxies in a noisy astronomical image. Both of the examples demonstrate phase transitions in the relationship between the likelihood and the cumulative prior mass, highlighting the need for Nested Sampling.
Cite
@article{arxiv.1411.3921,
title = {Inference for Trans-dimensional Bayesian Models with Diffusive Nested Sampling},
author = {Brendon J. Brewer},
journal= {arXiv preprint arXiv:1411.3921},
year = {2015}
}
Comments
Only published here for the time being. 17 pages, 10 figures. Software available at https://github.com/eggplantbren/RJObject