Constructing summary statistics for approximate Bayesian computation: semi‐automatic approximate Bayesian computation

By Paul Fearnhead, Dennis Prangle. 2012

629 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/04e6a87763fcdd2eb66f8d0f31d1ba4a5d5549da

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

Statistical inference for dynamical systems: A review (2012, background): “Choosing the metric and the tolerance level are difficult problems, but partial results are known ([35]).”
Dynamic Likelihood-free Inference via Ratio Estimation (DIRE) (2018, methodology): “ log(r) 2[3;5], ˙2 [0;0:6] and ˚2[5;15]. We compared our results with two other (nonLFIRE) inference methods | synthetic likelihood with the summary statistics by Wood (2010) and semiautomatic ABC by Fearnhead and Prangle (2012). The comparison is done in the same way as before for the Lotka-Volterr”
Bayesian computation: a summary of the current state, and samples backwards and forwards (2015, background): “…who links ABC with standard kernel density non-parametric estimation and ﬁnd a tolerance (re-expressed as a bandwidth) of order n − 1 / q + 4 and an rmse of order 2 / q + 4 as well, while Fearnhead and Prangle (2012) obtain similar rates, with a tolerance of order n − 1 / q + 2 for noisy ABC.”
Reliable ABC model choice via random forests (2014, methodology): “Other projection techniques have been proposed in the context of parameter estimation: see, e.g. Fearnhead and Prangle (2012); Aeschbacher et al. (2012).”
A comparative review of dimension reduction methods in approximate Bayesian computation (2012, methodology): “For example, in one application, Fearnhead and Prangle (2012) use the polynomial basis functions f(s) = (s, s2, s3, s4); that is, a vector of length 4p, where p = dim(s) is the number of elements in s, consisting of the first four powers of each element of s.”
Likelihood-free MCMC with Amortized Approximate Ratio Estimators (2019, background): “Other works investigated learning summary statistics (Fearnhead & Prangle, 2012; Dinev & Gutmann, 2018; Wong et al., 2018).”
Automatic Posterior Transformation for Likelihood-Free Inference (2019, methodology): “A central problem in ABC is the fact that virtually all algorithms rely on summary statistics in order to reduce the dimensionality of the data (Fearnhead & Prangle, 2012; Blum et al., 2013; Jiang et al., 2017)—designing summary statistics can require domain-speciﬁc knowledge and/or separate…”
Kernel approximate Bayesian computation in population genetic inferences (2012, methodology): “Fearnhead and Prangle (2012) introduced noisy-ABC, which calibrates the estimator, 𝑃�𝜽 ∈ 𝜣�𝑃 ( 𝜽 ∈ 𝜣 | 𝒔 ) = 𝑝 , 𝒟� = 𝑝 for ∀𝚯 , where 𝑃 ( 𝜽 ∈ is the probability assigned by the posterior density obtained by ABC with tolerance .”
Approximate Bayesian computation via regression density estimation (2012, methodology): “…methodology using the above summary statistics (corresponding to conditional density estimation in 115 dimensions), as well as the “semi-automatic” summary statistics of Fearnhead and Prangle (2012), which provide one summary statistic per parameter (i.e. 6-dimensional density estimation).”
The Rate of Convergence for Approximate Bayesian Computation (2013, methodology): “Fearnhead and Prangle (2012) obtain the same exponent − 2 / ( q + 4) for the speciﬁc summary statistic S ( X ) = E ( θ | X ).”
Bayesian Optimization for Likelihood-Free Inference of Simulator-Based Statistical Models (2015, background): “One difficulty is the assessment of the discrepancy between the observed and simulated data (Joyce and Marjoram, 2008; Wegmann et al., 2009; Nunes and Balding, 2010; Fearnhead and Prangle, 2012; Aeschbacher et al., 2012; Gutmann et al., 2014).”
Likelihood-Free Inference by Ratio Estimation (2016, background): “While there are several works on summary statistics selection in the framework of approximate Bayesian computation (Aeschbacher et al., 2012; Fearnhead and Prangle, 2012; Blum et al., 2013; Gutmann et al., 2014; Marin et al., 2016; Gutmann et al., 2018; Jiang et al., 2018), there is comparably littl”