Delayed acceptance particle MCMC for exact inference in stochastic kinetic models

DOI10.1007/S11222-014-9469-XzbMATH Open1332.60117arXiv1401.4369OpenAlexW3098926670WikidataQ61434479 ScholiaQ61434479MaRDI QIDQ261048FDOQ261048

Authors: Andrew Golightly, Daniel A. Henderson, Chris Sherlock 

Publication date: 22 March 2016

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: Recently-proposed particle MCMC methods provide a flexible way of performing Bayesian inference for parameters governing stochastic kinetic models defined as Markov (jump) processes (MJPs). Each iteration of the scheme requires an estimate of the marginal likelihood calculated from the output of a sequential Monte Carlo scheme (also known as a particle filter). Consequently, the method can be extremely computationally intensive. We therefore aim to avoid most instances of the expensive likelihood calculation through use of a fast approximation. We consider two approximations: the chemical Langevin equation diffusion approximation (CLE) and the linear noise approximation (LNA). Either an estimate of the marginal likelihood under the CLE, or the tractable marginal likelihood under the LNA can be used to calculate a first step acceptance probability. Only if a proposal is accepted under the approximation do we then run a sequential Monte Carlo scheme to compute an estimate of the marginal likelihood under the true MJP and construct a second stage acceptance probability that permits exact (simulation based) inference for the MJP. We therefore avoid expensive calculations for proposals that are likely to be rejected. We illustrate the method by considering inference for parameters governing a Lotka-Volterra system, a model of gene expression and a simple epidemic process.

Full work available at URL: https://arxiv.org/abs/1401.4369

Recommendations

• Bayesian computation methods for inference in stochastic kinetic models
• Efficient \(\mathrm{SMC}^2\) schemes for stochastic kinetic models
• Bayesian inference for hybrid discrete-continuous stochastic kinetic models
• Moment closure based parameter inference of stochastic kinetic models
• Correlated pseudo-marginal schemes for time-discretised stochastic kinetic models

zbMATH Keywords

Markov jump processchemical Langevin equationdelayed acceptancelinear noise approximationparticle MCMC

Mathematics Subject Classification ID

Computational methods in Markov chains (60J22) Applications of continuous-time Markov processes on discrete state spaces (60J28) Applications of statistics to biology and medical sciences; meta analysis (62P10) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40) Problems related to evolution (92D15)

Cites Work

• The pseudo-marginal approach for efficient Monte Carlo computations
• Particle Markov Chain Monte Carlo Methods
• Title not available (Why is that?)
• Efficient implementation of Markov chain Monte Carlo when using an unbiased likelihood estimator
• Monte Carlo strategies in scientific computing
• Automatic Selection of Methods for Solving Stiff and Nonstiff Systems of Ordinary Differential Equations
• On the efficiency of pseudo-marginal random walk Metropolis algorithms
• Stochastic modelling for systems biology.
• Particle Markov chain Monte Carlo for efficient numerical simulation
• Title not available (Why is that?)
• On some properties of Markov chain Monte Carlo simulation methods based on the particle filter
• Bayesian inference for nonlinear multivariate diffusion models observed with error
• Bayesian inference for stochastic epidemic models with time-inhomogeneous removal rates
• Network epidemic models with two levels of mixing
• A hierarchy of approximations of the master equation scaled by a size parameter
• Stochastic processes in physics and chemistry.
• Markov chain Monte Carlo inference for Markov jump processes via the linear noise approximation
• Exact Filtering for Partially Observed Continuous Time Models
• Inference for reaction networks using the linear noise approximation
• Solutions of ordinary differential equations as limits of pure jump markov processes
• Bayesian Inference for Stochastic Kinetic Models Using a Diffusion Approximation

Cited In (30)

• Approximating optimal SMC proposal distributions in individual-based epidemic models
• Multifidelity multilevel Monte Carlo to accelerate approximate Bayesian parameter inference for partially observed stochastic processes
• Geometric ergodicity of Rao and Teh's algorithm for homogeneous Markov jump processes
• Moment closure based parameter inference of stochastic kinetic models
• Efficiency of delayed-acceptance random walk metropolis algorithms
• Bayesian inference for hybrid discrete-continuous stochastic kinetic models
• Delayed Acceptance ABC-SMC
• Accelerating pseudo-marginal MCMC using Gaussian processes
• Speeding up MCMC by Delayed Acceptance and Data Subsampling
• Likelihood free inference for Markov processes: a comparison
• A linear noise approximation for stochastic epidemic models fit to partially observed incidence counts
• Ensemble MCMC: accelerating pseudo-marginal MCMC for state space models using the ensemble Kalman filter
• Inference for stochastic kinetic models from multiple data sources for joint estimation of infection dynamics from aggregate reports and virological data
• A Function Emulation Approach for Doubly Intractable Distributions
• Variance bounding of delayed-acceptance kernels
• Accelerating inference for stochastic kinetic models
• Bayesian inference for Markov jump processes with informative observations
• Rapid Bayesian Inference for Expensive Stochastic Models
• Efficient sequential Monte Carlo algorithms for integrated population models
• Importance sampling correction versus standard averages of reversible MCMCs in terms of the asymptotic variance
• Markov chain Monte Carlo inference for Markov jump processes via the linear noise approximation
• Accelerating sequential Monte Carlo with surrogate likelihoods
• Improved bridge constructs for stochastic differential equations
• Emulation-accelerated Hamiltonian Monte Carlo algorithms for parameter estimation and uncertainty quantification in differential equation models
• Scalable inference for Markov processes with intractable likelihoods
• Computing Bayes: from then `til now
• An algorithm for approximating the second moment of the normalizing constant estimate from a particle filter
• Diagnostics for assessing the linear noise and moment closure approximations
• Bayesian computation methods for inference in stochastic kinetic models
• Expectation propagation for continuous time stochastic processes

This page was built for publication: Delayed acceptance particle MCMC for exact inference in stochastic kinetic models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q261048)