Likelihood-free stochastic approximation EM for inference in complex models
arXiv:1609.03508 · doi:10.1080/03610918.2017.1401082
Abstract
A maximum likelihood methodology for the parameters of models with an intractable likelihood is introduced. We produce a likelihood-free version of the stochastic approximation expectation-maximization (SAEM) algorithm to maximize the likelihood function of model parameters. While SAEM is best suited for models having a tractable "complete likelihood" function, its application to moderately complex models is a difficult or even impossible task. We show how to construct a likelihood-free version of SAEM by using the "synthetic likelihood" paradigm. Our method is completely plug-and-play, requires almost no tuning and can be applied to both static and dynamic models. Four simulation studies illustrate the method, including a stochastic differential equation model, a stochastic Lotka-Volterra model and data from -and- distributions. MATLAB code is available as supplementary material.
arXiv commentFixed a couple of typos, e.g. in algorithm 3, step (i), we now have y*~p(Y|x*) instead of y*~p(Y|x). Similarly on page 7 (step 1 of Internal SAEM-SL). Published in "Communications in Statistics: Simulation and Computation" doi:10.1080/03610918.2017.1401082
References in corpus (9)
- Approximate Bayesian computation scheme for parameter inference and model selection in dynamical systems
- The pseudo-marginal approach for efficient Monte Carlo computations
- On Particle Methods for Parameter Estimation in State-Space Models
- On the efficiency of pseudo-marginal random walk Metropolis algorithms
- Fast -free Inference of Simulation Models with Bayesian Conditional Density Estimation
- Inference for SDE models via Approximate Bayesian Computation
- Estimation in the partially observed stochastic Morris-Lecar neuronal model with particle filter and stochastic approximation methods
- Coupling stochastic EM and Approximate Bayesian Computation for parameter inference in state-space models
- Approximate maximum likelihood estimation using data-cloning ABC