Automated Parameter Blocking for Efficient Markov-Chain Monte Carlo Sampling

Markov chain Monte Carlo (MCMC) sampling is an important and commonly used tool for the analysis of hierarchical models. Nevertheless, practitioners generally have two options for MCMC: utilize existing software that generates a black-box "one size fits all" algorithm, or the challenging (and time consuming) task of implementing a problem-specific MCMC algorithm. Either choice may result in inefficient sampling, and hence researchers have become accustomed to MCMC runtimes on the order of days (or longer) for large models. We propose an automated procedure to determine an efficient MCMC algorithm for a given model and computing platform. Our procedure dynamically determines blocks of parameters for joint sampling that result in efficient sampling of the entire model. We test this procedure using a diverse suite of example models, and observe non-trivial improvements in MCMC efficiency for many models. Our procedure is the first attempt at such, and may be generalized to a broader space of MCMC algorithms. Our results suggest that substantive improvements in MCMC efficiency may be practically realized using our automated blocking procedure, or variants thereof, which warrants additional study and application.

[1]  Hans J. Skaug,et al.  Automatic approximation of the marginal likelihood in non-Gaussian hierarchical models , 2006, Comput. Stat. Data Anal..

[2]  L A Waller,et al.  Log-linear modeling with the negative multinomial distribution. , 1997, Biometrics.

[3]  M. Plummer,et al.  CODA: convergence diagnosis and output analysis for MCMC , 2006 .

[4]  Heikki Haario,et al.  Adaptive proposal distribution for random walk Metropolis algorithm , 1999, Comput. Stat..

[5]  Galin L. Jones,et al.  Ascent‐based Monte Carlo expectation– maximization , 2005 .

[6]  N. Breslow,et al.  Approximate inference in generalized linear mixed models , 1993 .

[7]  Andrew Gelman,et al.  Data Analysis Using Regression and Multilevel/Hierarchical Models , 2006 .

[8]  N. Hjort,et al.  Post-Processing Posterior Predictive p Values , 2006 .

[9]  Jun S. Liu,et al.  Covariance structure of the Gibbs sampler with applications to the comparisons of estimators and augmentation schemes , 1994 .

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

[11]  Gareth Roberts,et al.  An adaptive approach to Langevin MCMC , 2012, Stat. Comput..

[12]  Madeleine B. Thompson.  A Comparison of Methods for Computing Autocorrelation Time , 2010, 1011.0175.

[13]  Weidong Liu,et al.  Adaptive Thresholding for Sparse Covariance Matrix Estimation , 2011, 1102.2237.

[14]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[15]  Gareth O. Roberts,et al.  Examples of Adaptive MCMC , 2009 .

[16]  R. Tibshirani,et al.  An introduction to the bootstrap , 1993 .

[17]  Leonhard Held,et al.  Gaussian Markov Random Fields: Theory and Applications , 2005 .

[18]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[19]  James Durbin,et al.  Time Series Analysis by State Space Methods: Second Edition , 2012 .

[20]  Han Lin Shang,et al.  The BUGS book: a practical introduction to Bayesian analysis , 2013 .

[21]  Neville Davies,et al.  Time Series Models, 2nd Edn. , 1995 .

[22]  R. Tweedie,et al.  Geometric convergence and central limit theorems for multidimensional Hastings and Metropolis algorithms , 1996 .

[23]  Sw. Banerjee,et al.  Hierarchical Modeling and Analysis for Spatial Data , 2003 .

[24]  Brian Dennis,et al.  Data cloning: easy maximum likelihood estimation for complex ecological models using Bayesian Markov chain Monte Carlo methods. , 2007, Ecology letters.

[25]  Radford M. Neal.  MCMC Using Hamiltonian Dynamics , 2011, 1206.1901.

[26]  Andrew Thomas,et al.  WinBUGS - A Bayesian modelling framework: Concepts, structure, and extensibility , 2000, Stat. Comput..

[27]  A. Raftery,et al.  Strictly Proper Scoring Rules, Prediction, and Estimation , 2007 .

[28]  R. Tweedie,et al.  Rates of convergence of the Hastings and Metropolis algorithms , 1996 .

[29]  S. Duane,et al.  Hybrid Monte Carlo , 1987 .

[30]  L. Trefethen,et al.  Numerical linear algebra , 1997 .

[31]  Martyn Plummer,et al.  JAGS Version 3.3.0 user manual , 2012 .

[32]  G. Roberts,et al.  Updating Schemes, Correlation Structure, Blocking and Parameterization for the Gibbs Sampler , 1997 .

[33]  James G. Booth,et al.  Monte Carlo EM for Generalized Linear Mixed Models using Randomized Spherical Radial Integration , 2006 .

[34]  ALGORITHMSG,et al.  WEAK CONVERGENCE AND OPTIMAL SCALINGOF RANDOM WALK METROPOLIS , 1994 .

[35]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[36]  A. J. Stam,et al.  Estimation of statistical errors in molecular simulation calculations , 1986 .

[37]  A. Gelman,et al.  Weak convergence and optimal scaling of random walk Metropolis algorithms , 1997 .

[38]  Radford M. Neal.  Slice Sampling , 2003, The Annals of Statistics.

[39]  Bradley P. Carlin,et al.  Structured Markov Chain Monte Carlo , 2000 .

[40]  Mahesh Motwani,et al.  Survey of clustering algorithms for MANET , 2009, ArXiv.