Particle gibbs with ancestor sampling

Particle Markov chain Monte Carlo (PMCMC) is a systematic way of combining the two main tools used for Monte Carlo statistical inference: sequential Monte Carlo (SMC) and Markov chain Monte Carlo (MCMC). We present a new PMCMC algorithm that we refer to as particle Gibbs with ancestor sampling (PGAS). PGAS provides the data analyst with an off-the-shelf class of Markov kernels that can be used to simulate, for instance, the typically high-dimensional and highly autocorrelated state trajectory in a state-space model. The ancestor sampling procedure enables fast mixing of the PGAS kernel even when using seemingly few particles in the underlying SMC sampler. This is important as it can significantly reduce the computational burden that is typically associated with using SMC. PGAS is conceptually similar to the existing PG with backward simulation (PGBS) procedure. Instead of using separate forward and backward sweeps as in PGBS, however, we achieve the same effect in a single forward sweep. This makes PGAS well suited for addressing inference problems not only in state-space models, but also in models with more complex dependencies, such as non-Markovian, Bayesian nonparametric, and general probabilistic graphical models.

[1]  M. Pitt,et al.  Filtering via Simulation: Auxiliary Particle Filters , 1999 .

[2]  A. Doucet,et al.  A Tutorial on Particle Filtering and Smoothing: Fifteen years later , 2008 .

[3]  Jasper A. Vrugt,et al.  Hydrologic data assimilation using particle Markov chain Monte Carlo simulation: Theory, concepts and applications (online first) , 2012 .

[4]  Gareth O. Roberts,et al.  Non-centred parameterisations for hierarchical models and data augmentation. , 2003 .

[5]  Charles J. Geyer,et al.  Practical Markov Chain Monte Carlo , 1992 .

[6]  C. Andrieu,et al.  Markovian stochastic approximation with expanding projections , 2011, 1111.5421.

[7]  Carl E. Rasmussen,et al.  State-Space Inference and Learning with Gaussian Processes , 2010, AISTATS.

[8]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[9]  M. Keeling,et al.  Modeling Infectious Diseases in Humans and Animals , 2007 .

[10]  Darren J Wilkinson,et al.  Bayesian parameter inference for stochastic biochemical network models using particle Markov chain Monte Carlo , 2011, Interface Focus.

[11]  Darren J. Wilkinson,et al.  Bayesian inference for nonlinear multivariate diffusion models observed with error , 2008, Comput. Stat. Data Anal..

[12]  M. Chesney,et al.  Pricing European Currency Options: A Comparison of the Modified Black-Scholes Model and a Random Variance Model , 1989, Journal of Financial and Quantitative Analysis.

[13]  Pierre Priouret,et al.  Adaptive Algorithms and Stochastic Approximations , 1990, Applications of Mathematics.

[14]  G. C. Wei,et al.  A Monte Carlo Implementation of the EM Algorithm and the Poor Man's Data Augmentation Algorithms , 1990 .

[15]  Branko Ristic,et al.  Monitoring and prediction of an epidemic outbreak using syndromic observations. , 2011, Mathematical biosciences.

[16]  Branko Ristic,et al.  Beyond the Kalman Filter: Particle Filters for Tracking Applications , 2004 .

[17]  Fredrik Lindsten,et al.  Ancestor Sampling for Particle Gibbs , 2012, NIPS.

[18]  É. Moulines,et al.  Convergence of a stochastic approximation version of the EM algorithm , 1999 .

[19]  Hoon Kim,et al.  Monte Carlo Statistical Methods , 2000, Technometrics.

[20]  P. Fearnhead,et al.  Particle filters for partially observed diffusions , 2007, 0710.4245.

[21]  Christophe Andrieu,et al.  Uniform ergodicity of the iterated conditional SMC and geometric ergodicity of particle Gibbs samplers , 2013, 1312.6432.

[22]  Tim Hesterberg,et al.  Monte Carlo Strategies in Scientific Computing , 2002, Technometrics.

[23]  D. V. van Dyk,et al.  Partially Collapsed Gibbs Samplers , 2008 .

[24]  A. Doucet,et al.  Monte Carlo Smoothing for Nonlinear Time Series , 2004, Journal of the American Statistical Association.

[25]  Andreas Krause,et al.  Advances in Neural Information Processing Systems (NIPS) , 2014 .

[26]  Christian P. Robert,et al.  Monte Carlo Statistical Methods (Springer Texts in Statistics) , 2005 .

[27]  N. Shephard,et al.  The simulation smoother for time series models , 1995 .

[28]  D. Blei Bayesian Nonparametrics I , 2016 .

[29]  Sophie Donnet,et al.  EM algorithm coupled with particle filter for maximum likelihood parameter estimation of stochastic differential mixed-effects models , 2010 .

[30]  Fredrik Lindsten,et al.  Backward Simulation Methods for Monte Carlo Statistical Inference , 2013, Found. Trends Mach. Learn..

[31]  Tobias Rydén,et al.  Rao-Blackwellization of Particle Markov Chain Monte Carlo Methods Using Forward Filtering Backward Sampling , 2011, IEEE Transactions on Signal Processing.

[32]  P. Müller,et al.  Bayesian Nonparametrics: An invitation to Bayesian nonparametrics , 2010 .

[33]  Michael I. Jordan,et al.  Hierarchical Dirichlet Processes , 2006 .

[34]  David A. Rasmussen,et al.  Inference for Nonlinear Epidemiological Models Using Genealogies and Time Series , 2011, PLoS Comput. Biol..

[35]  A. Doucet,et al.  Particle Markov chain Monte Carlo methods , 2010 .

[36]  Sumeetpal S. Singh,et al.  On the Particle Gibbs Sampler , 2013 .

[37]  Fredrik Gustafsson,et al.  Particle filters for positioning, navigation, and tracking , 2002, IEEE Trans. Signal Process..

[38]  Sumeetpal S. Singh,et al.  On particle Gibbs sampling , 2013, 1304.1887.

[39]  M. J. Bayarri,et al.  Non-Centered Parameterisations for Hierarchical Models and Data Augmentation , 2003 .

[40]  D. Stephens,et al.  Stochastic volatility modelling in continuous time with general marginal distributions: Inference, prediction and model selection , 2007 .

[41]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[42]  Ralph S. Silva,et al.  On Some Properties of Markov Chain Monte Carlo Simulation Methods Based on the Particle Filter , 2012 .

[43]  Linda C. van der Gaag,et al.  Probabilistic Graphical Models , 2014, Lecture Notes in Computer Science.

[44]  A. Doucet,et al.  Efficient implementation of Markov chain Monte Carlo when using an unbiased likelihood estimator , 2012, 1210.1871.

[45]  Fredrik Lindsten,et al.  On the use of backward simulation in the particle Gibbs sampler , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[46]  Jun S. Liu Peskun's theorem and a modified discrete-state Gibbs sampler , 1996 .

[47]  L. Tierney Markov Chains for Exploring Posterior Distributions , 1994 .

[48]  M. Escobar,et al.  Bayesian Density Estimation and Inference Using Mixtures , 1995 .

[49]  S. Turnbull,et al.  Pricing foreign currency options with stochastic volatility , 1990 .

[50]  G. Bierman Fixed interval smoothing with discrete measurements , 1972 .

[51]  Jeremy Ginsberg,et al.  Detecting influenza epidemics using search engine query data , 2009, Nature.

[52]  M. Li,et al.  Particle Markov chain Monte Carlo methods , 2015 .

[53]  Simon J. Godsill,et al.  Rao-Blackwellized particle smoothers for mixed linear/nonlinear state-space models , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[54]  P. Moral,et al.  Sequential Monte Carlo samplers , 2002, cond-mat/0212648.

[55]  N. Shephard,et al.  Stochastic Volatility: Likelihood Inference And Comparison With Arch Models , 1996 .

[56]  Giuseppe Cavaliere Stochastic Volatility: Selected Readings , 2006 .

[57]  A. Doucet,et al.  Efficient Bayesian Inference for Switching State-Space Models using Discrete Particle Markov Chain Monte Carlo Methods , 2010, 1011.2437.

[58]  Jun S. Liu,et al.  Mixture Kalman filters , 2000 .

[59]  Pierre E. Jacob,et al.  Path storage in the particle filter , 2013, Statistics and Computing.

[60]  Darren J. Wilkinson,et al.  Parallel Bayesian Computation , 2005 .

[61]  Carl E. Rasmussen,et al.  Bayesian Inference and Learning in Gaussian Process State-Space Models with Particle MCMC , 2013, NIPS.

[62]  John Parslow,et al.  On Disturbance State-Space Models and the Particle Marginal Metropolis-Hastings Sampler , 2012, SIAM/ASA J. Uncertain. Quantification.

[63]  Timo Koski,et al.  Decomposition Sampling applied to Parallelization of Metropolis-Hastings , 2015 .

[64]  Eric Moulines,et al.  Stability of Stochastic Approximation under Verifiable Conditions , 2005, Proceedings of the 44th IEEE Conference on Decision and Control.

[65]  Fredrik Lindsten,et al.  An efficient stochastic approximation EM algorithm using conditional particle filters , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[66]  Robert Kohn,et al.  Auxiliary particle ltering within adaptive Metropolis-Hastings sampling , 2010, 1006.1914.