Bayesian model comparison with un-normalised likelihoods

详细信息    查看全文

• 作者：Richard G. Everitt ; Adam M. Johansen ; Ellen Rowing…
• 关键词：Approximate Bayesian computation ; Bayes’ factors ; Importance sampling ; Marginal likelihood ; Markov random field ; Partition function ; Sequential Monte Carlo
• 刊名：Statistics and Computing
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：27
• 期：2
• 页码：403-422
• 全文大小：
• 刊物类别：Mathematics and Statistics
• 刊物主题：Statistics and Computing/Statistics Programs; Artificial Intelligence (incl. Robotics); Statistical Theory and Methods; Probability and Statistics in Computer Science;
• 出版者：Springer US
• ISSN：1573-1375
• 卷排序：27

文摘

Models for which the likelihood function can be evaluated only up to a parameter-dependent unknown normalizing constant, such as Markov random field models, are used widely in computer science, statistical physics, spatial statistics, and network analysis. However, Bayesian analysis of these models using standard Monte Carlo methods is not possible due to the intractability of their likelihood functions. Several methods that permit exact, or close to exact, simulation from the posterior distribution have recently been developed. However, estimating the evidence and Bayes’ factors for these models remains challenging in general. This paper describes new random weight importance sampling and sequential Monte Carlo methods for estimating BFs that use simulation to circumvent the evaluation of the intractable likelihood, and compares them to existing methods. In some cases we observe an advantage in the use of biased weight estimates. An initial investigation into the theoretical and empirical properties of this class of methods is presented. Some support for the use of biased estimates is presented, but we advocate caution in the use of such estimates.