Likelihood-free Markov chain Monte Carlo

By Scott Anthony Sisson, Yanan Fan. 2010

112 citations recorded by the source corpus.

To appear to MCMC handbook, S. P. Brooks, A. Gelman, G. Jones and X.-L. Meng (eds), Chapman & Hall.

Subfields: Artificial Intelligence, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

Gradient-free Hamiltonian Monte Carlo with Efficient Kernel Exponential Families (2015, methodology): “A related setting is MCMC for Approximate Bayesian Computation (ABC-MCMC), where the posterior is approximated through repeated simulation from a likelihood model [5, 6].”
Matching, sorting and wages (2011, methodology): “The moments we choose are the transitions between employment states, the level of wages and their cross sectional variance one year, eleven years and twentyone years following labor market (5)Details pertaining to tuning the MCMC algorithm, a parallel implementation, and related methods in statistic”
GPS-ABC: Gaussian Process Surrogate Approximate Bayesian Computation (2014, methodology): “The current state-of-the-art for simulation-based model inference is likelihood-free or approximate Bayesian computation methods [23, 15].”
A comparative review of dimension reduction methods in approximate Bayesian computation (2012, methodology): “A toy example illustrating this was given by Sisson and Fan (2011).”
On-the-Job Search and Precautionary Savings (2013, methodology): “Details pertaining to tuning the MCMC algorithm, a parallel implementation, and related methods in statistics can be found in Robert and Casella (2004), Vrugt et al. (2009), and Sisson and Fan (2011).”
On sequential Monte Carlo, partial rejection control and approximate Bayesian computation (2008, unknown): “When these methods are non-exact, in that they implement an approximation to the intractable posterior, they are commonly termed approximate Bayesian computation (ABC) methods (e.g. Csilleŕy et al. 2010; Sisson and Fan 2011).”
Approximate Bayesian computation via regression density estimation (2012, methodology): “…et al. 2007; Ratmann et al. 2009), ABC has now found application in a wide range of areas – see the reviews by Beaumont (2010), Csilléry et al. (2010) and Sisson and Fan (2011) ABC methods require the ability to simulate data from the intractable likelihood, L(·|θ), for a given parameter vector θ.”
Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows (2018, methodology): “A basic implementation of ABC would simulate parameters from the prior; more sophisticated variants such as Markov Chain Monte Carlo ABC [47, 51, 69] and Sequential Monte Carlo ABC [6, 10, 70, 76] guide future simulations based on previously accepted parameters.”
Hamiltonian ABC (2015, methodology): “…eq 2, we arrive at the following Metropolis-Hastings accept-reject probability, If the simulations are part of the Markov chain, the algorithm corresponds to the pseudo-marginal (PM) sampler (Andrieu & Roberts, 2009), otherwise it is a marginal sampler (Marjoram et al., 2003; Sisson & Fan, 2010).”
Likelihood-free MCMC with Approximate Likelihood Ratios (2019, methodology): “These samples are in turn evaluated by the Metropolis-mechanism under a summary statistic [34, 49].”