Projects per year
Abstract
A maximum likelihood methodology for the parameters of models with an intractable likelihood is introduced. We produce a likelihoodfree version of the stochastic approximation expectationmaximization (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 likelihoodfree version of SAEM by using the "synthetic likelihood" paradigm. Our method is completely plugandplay, 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 LotkaVolterra model and data from gandk distributions. MATLAB code is available as supplementary material.
Original language  English 

Pages (fromto)  861881 
Number of pages  26 
Journal  Communications in Statistics: Simulation and Computation 
Volume  48 
Issue number  3 
Early online date  2018 Jan 18 
DOIs  
Publication status  Published  2019 
Subject classification (UKÄ)
 Probability Theory and Statistics
Keywords
 maximum likelihood
 SAEM
 sequential Monte Carlo
 synthetic likelihood;
 state space model
 Stochastic differential equation
Fingerprint
Dive into the research topics of 'Likelihoodfree stochastic approximation EM for inference in complex models'. Together they form a unique fingerprint.Projects
 1 Active

Stochastic modelling of protein folding and likelihoodfree statistical inference methods
Picchini, U., Forman, J., LindorffLarsen, K. & Wiqvist, S.
2015/01/01 → …
Project: Research