On predictive inference for intractable models via approximate Bayesian computation
From MaRDI portal
Abstract: Approximate Bayesian computation (ABC) is commonly used for parameter estimation and model comparison for intractable simulator-based models whose likelihood function cannot be evaluated. In this paper we instead investigate the feasibility of ABC as a generic approximate method for predictive inference, in particular, for computing the posterior predictive distribution of future observations or missing data of interest. We consider three complementary ABC approaches for this goal, each based on different assumptions regarding which predictive density of the intractable model can be sampled from. The case where only simulation from the joint density of the observed and future data given the model parameters can be used for inference is given particular attention and it is shown that the ideal summary statistic in this setting is minimal predictive sufficient instead of merely minimal sufficient (in the ordinary sense). An ABC prediction approach that takes advantage of a certain latent variable representation is also investigated. We additionally show how common ABC sampling algorithms can be used in the predictive settings considered. Our main results are first illustrated by using simple time-series models that facilitate analytical treatment, and later by using two common intractable dynamic models.
Recommendations
- Constructing summary statistics for approximate Bayesian computation: semi-automatic approximate Bayesian computation. With discussion and authors' reply
- An Introduction to Approximate Bayesian Computation
- New insights into approximate Bayesian computation
- Asymptotic properties of approximate Bayesian computation
- A tutorial on approximate Bayesian computation
Cites work
- A comparison of inferential methods for highly nonlinear state space models in ecology and epidemiology
- A survey of Bayesian predictive methods for model assessment, selection and comparison
- A tutorial introduction to Bayesian inference for stochastic epidemic models using approximate Bayesian computation
- Adapting the ABC distance function
- Adaptive approximate Bayesian computation
- Adaptive approximate Bayesian computation tolerance selection
- Adequate Subfields and Sufficiency
- An adaptive sequential Monte Carlo method for approximate Bayesian computation
- Approximate Bayesian Computation for a Class of Time Series Models
- Approximate Bayesian Computation for Smoothing
- Approximate Bayesian Computation with the Wasserstein Distance
- Approximate Bayesian computation: a nonparametric perspective
- Approximate Bayesian computational methods
- Approximate Bayesian inference for discretely observed continuous‐time multi‐state models
- Approximate leave-future-out cross-validation for Bayesian time series models
- Auxiliary Likelihood-Based Approximate Bayesian Computation in State Space Models
- Bayesian nonparametric forecasting of monotonic functional time series
- Bayesian Synthetic Likelihood
- Constructing summary statistics for approximate Bayesian computation: semi-automatic approximate Bayesian computation. With discussion and authors' reply
- Efficient sampling of conditioned Markov jump processes
- Estimating Functions in Indirect Inference
- Estimating the transmission dynamics of \textit{Streptococcus pneumoniae} from strain prevalence data
- Filtering and estimation for a class of stochastic volatility models with intractable likelihoods
- Filtering via approximate Bayesian computation
- Handbook of approximate Bayesian computation
- scientific article; zbMATH DE number 3466390 (Why is no real title available?)
- scientific article; zbMATH DE number 2171466 (Why is no real title available?)
- scientific article; zbMATH DE number 947414 (Why is no real title available?)
- scientific article; zbMATH DE number 1865509 (Why is no real title available?)
- New insights into approximate Bayesian computation
- On the Generalization of the Likelihood Function and the Likelihood Principle
- Particle Markov Chain Monte Carlo Methods
- Sequential Monte Carlo without likelihoods
- Spectral density-based and measure-preserving ABC for partially observed diffusion processes. An illustration on Hamiltonian SDEs
- Stochastic modelling for systems biology
- The rate of convergence for approximate Bayesian computation
- Theory of statistics
Cited in
(2)
This page was built for publication: On predictive inference for intractable models via approximate Bayesian computation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6171773)
