Delayed Acceptance ABC-SMC
Open Access
- 9 July 2020
- journal article
- research article
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 30 (1) , 55-66
Abstract
Approximate Bayesian computation (ABC) is now an established technique for statistical inference used in cases where the likelihood function is computationally expensive or not available. It relies on the use of a model that is specified in the form of a simulator, and approximates the likelihood at a parameter value θ by simulating auxiliary data sets x and evaluating the distance of x from the true data y. However, ABC is not computationally feasible in cases where using the simulator for each θ is very expensive. This paper investigates this situation in cases where a cheap, but approximate, simulator is available. The approach is to employ delayed acceptance Markov chain Monte Carlo (MCMC) within an ABC sequential Monte Carlo (SMC) sampler in order to, in a first stage of the kernel, use the cheap simulator to rule out parts of the parameter space that are not worth exploring, so that the “true” simulator is only run (in the second stage of the kernel) where there is a reasonable chance of accepting proposed values of θ. We show that this approach can be used quite automatically, with few tuning parameters. Applications to stochastic differential equation models and latent doubly intractable distributions are presented.Keywords
All Related Versions
Funding Information
- BBSRC (BB/N00874X/1)
- EPSRC (EP/N023927/1)
This publication has 31 references indexed in Scilit:
- On adaptive resampling strategies for sequential Monte Carlo methodsBernoulli, 2012
- Likelihood-free estimation of model evidenceBayesian Analysis, 2011
- ABC likelihood-free methods for model choice in Gibbs random fieldsBayesian Analysis, 2009
- The pseudo-marginal approach for efficient Monte Carlo computationsThe Annals of Statistics, 2009
- Sequential Monte Carlo without likelihoodsProceedings of the National Academy of Sciences, 2007
- An efficient Markov chain Monte Carlo method for distributions with intractable normalising constantsBiometrika, 2006
- Sequential Monte Carlo SamplersJournal of the Royal Statistical Society Series B: Statistical Methodology, 2006
- A note on Metropolis-Hastings kernels for general state spacesThe Annals of Applied Probability, 1998
- Sequential Imputations and Bayesian Missing Data ProblemsJournal of the American Statistical Association, 1994
- Exact stochastic simulation of coupled chemical reactionsThe Journal of Physical Chemistry, 1977