Likelihood free inference for Markov processes: a comparison
From MaRDI portal
Publication:2344258
DOI10.1515/SAGMB-2014-0072zbMATH Open1311.60079arXiv1410.0524OpenAlexW1987625453WikidataQ46767208 ScholiaQ46767208MaRDI QIDQ2344258FDOQ2344258
Authors: J. Owen, Colin S. Gillespie, Darren J. Wilkinson 
Publication date: 13 May 2015
Published in: Statistical Applications in Genetics and Molecular Biology (Search for Journal in Brave)
Abstract: Approaches to Bayesian inference for problems with intractable likelihoods have become increasingly important in recent years. Approximate Bayesian computation (ABC) and "likelihood free" Markov chain Monte Carlo techniques are popular methods for tackling inference in these scenarios but such techniques are computationally expensive. In this paper we compare the two approaches to inference, with a particular focus on parameter inference for stochastic kinetic models, widely used in systems biology. Discrete time transition kernels for models of this type are intractable for all but the most trivial systems yet forward simulation is usually straightforward. We discuss the relative merits and drawbacks of each approach whilst considering the computational cost implications and efficiency of these techniques. In order to explore the properties of each approach we examine a range of observation regimes using two example models. We use a Lotka--Volterra predator prey model to explore the impact of full or partial species observations using various time course observations under the assumption of known and unknown measurement error. Further investigation into the impact of observation error is then made using a Schl"ogl system, a test case which exhibits bi-modal state stability in some regions of parameter space.
Full work available at URL: https://arxiv.org/abs/1410.0524
Recommendations
- Likelihood-free MCMC
- Likelihood-free estimation of model evidence
- Delayed acceptance particle MCMC for exact inference in stochastic kinetic models
- Bayesian computation methods for inference in stochastic kinetic models
- Piecewise approximate Bayesian computation: fast inference for discretely observed Markov models using a factorised posterior distribution
Bayesian inference (62F15) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Population dynamics (general) (92D25)
Cites Work
- Sequential Monte Carlo Methods in Practice
- Sequential Monte Carlo Samplers
- Weak convergence and optimal scaling of random walk Metropolis algorithms
- The pseudo-marginal approach for efficient Monte Carlo computations
- A comparative review of dimension reduction methods in approximate Bayesian computation
- Constructing Summary Statistics for Approximate Bayesian Computation: Semi-Automatic Approximate Bayesian Computation
- Efficient implementation of Markov chain Monte Carlo when using an unbiased likelihood estimator
- Sequential Monte Carlo without likelihoods
- Scalable inference for Markov processes with intractable likelihoods
- Optimal scaling for various Metropolis-Hastings algorithms.
- Adaptive approximate Bayesian computation
- On the efficiency of pseudo-marginal random walk Metropolis algorithms
- Estimation of parameters for macroparasite population evolution using approximate Bayesian computation
- Particle Markov chain Monte Carlo for efficient numerical simulation
- On some properties of Markov chain Monte Carlo simulation methods based on the particle filter
- Bayesian Inference for Stochastic Kinetic Models Using a Diffusion Approximation
- Bayesian inference for Markov jump processes with informative observations
- Convergence of adaptive mixtures of importance sampling schemes
Cited In (9)
- A comparison of inferential methods for highly nonlinear state space models in ecology and epidemiology
- Approximate maximum likelihood estimation using data-cloning ABC
- Direct statistical inference for finite Markov jump processes via the matrix exponential
- Correlated pseudo-marginal schemes for time-discretised stochastic kinetic models
- Efficient parametric inference for stochastic biological systems with measured variability
- Accelerating inference for stochastic kinetic models
- Approximate Bayesian computation in controlled branching processes: the role of summary statistics
- Diagnostics for assessing the linear noise and moment closure approximations
- Introduction to ``Scalable inference for Markov processes with intractable likelihoods by J. Owen, D. Wilkinson, C. Gillespie
This page was built for publication: Likelihood free inference for Markov processes: a comparison
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2344258)
