COSMOLOGICAL MODEL SELECTION

Abstract

We give an overview of the recent progress in the field of cosmological model selection. Model selection statistics, such as those based on information theory and on Bayesian statistics are introduced and discussed. In the Bayesian framework, the marginalised model likelihood, or evidence, is the primary model selection statistic. We describe different methods of computing the evidence, and focus in particular on Nested Sampling. We describe the results of applying model selection methods to new cosmological data such as the CMB measurements by WMAP.