Abstract
We provide a gentle introduction to the particle Metropolis-Hastings (PMH)
algorithm for parameter inference in nonlinear state space models (SSMs)
together with a software implementation in the statistical programming language
R. Throughout this tutorial, we develop an implementation of the PMH algorithm
(and the integrated particle filter), which is distributed as the package
pmhtutorial available from the CRAN repository. Moreover, we provide the reader
with some intuition for how the algorithm operates and discuss some solutions
to numerical problems that might occur in practice. To illustrate the use of
PMH, we consider parameter inference in a linear Gaussian SSM with synthetic
data and a nonlinear stochastic volatility model with real-world data. We
conclude the tutorial by discussing important possible improvements to the
algorithm and we also list suitable references for further study.
Users
Please
log in to take part in the discussion (add own reviews or comments).