Cross‐validation based assessment of a new Bayesian palaeoclimate model

Fossil-based palaeoclimate reconstruction is an important area of ecological science that has gained momentum in the backdrop of the global climate change debate. The hierarchical Bayesian paradigm provides an interesting platform for studying such important scientific issue. However, our cross-validation based assessment of the existing Bayesian hierarchical models with respect to two modern proxy data sets based on chironomid and pollen, respectively, revealed that the models are inadequate for the data sets. In this paper, we model the species assemblages (compositional data) by the zero-inflated multinomial distribution while modelling the species response functions using Dirichlet process-based Gaussian mixtures. This modelling strategy yielded significantly improved performances, and a formal Bayesian test of model adequacy, developed recently, showed that our new model is adequate for both the modern data sets. Furthermore, combining together the zero-inflated assumption, Importance Resampling Markov Chain Monte Carlo (IRMCMC) and the recently developed Transformation-based Markov Chain Monte Carlo (TMCMC), we develop a powerful and efficient computational methodology. Copyright © 2013 John Wiley & Sons, Ltd.

[1]  Hannu Toivonen,et al.  Holocene temperature changes in northern Fennoscandia reconstructed from chironomids using Bayesian modelling , 2002 .

[2]  Lancelot F. James,et al.  Gibbs Sampling Methods for Stick-Breaking Priors , 2001 .

[3]  Hal Daumé,et al.  Fast search for Dirichlet process mixture models , 2007, AISTATS.

[4]  P. Green,et al.  Modelling Heterogeneity With and Without the Dirichlet Process , 2001 .

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

[6]  Michael Salter-Townshend,et al.  Fast inversion of a flexible regression model for multivariate pollen counts data , 2012 .

[7]  T. Ferguson.  A Bayesian Analysis of Some Nonparametric Problems , 1973 .

[8]  Adrian F. M. Smith,et al.  Automatic Bayesian curve fitting , 1998 .

[9]  R. Battarbee.  Palaeolimnological approaches to climate change, with special regard to the biological record , 2000 .

[10]  D. Ruppert.  Selecting the Number of Knots for Penalized Splines , 2002 .

[11]  Christian Ohlwein,et al.  Review of probabilistic pollen-climate transfer methods , 2012 .

[12]  Shane T. Jensen,et al.  Bayesian Clustering of Transcription Factor Binding Motifs , 2006, math/0610655.

[13]  S. Mukhopadhyay,et al.  Bayesian MISE Convergence Rates of Mixture Models Based on the Polya Urn Model: Asymptotic Comparisons and Choice of Prior Parameters , 2012 .

[14]  Simon P. Wilson,et al.  Bayesian palaeoclimate reconstruction , 2006 .

[15]  Paul Fearnhead,et al.  Particle filters for mixture models with an unknown number of components , 2004, Stat. Comput..

[16]  Radford M. Neal.  Markov Chain Sampling Methods for Dirichlet Process Mixture Models , 2000 .

[17]  Hannu Toivonen,et al.  A Bayesian multinomial Gaussian response model for organism-based environmental reconstruction , 2000 .

[18]  Lancelot F. James,et al.  Bayesian Model Selection in Finite Mixtures by Marginal Density Decompositions , 2001 .

[19]  D. B. Dahl.  Modal clustering in a class of product partition models , 2009 .

[20]  S. Mukhopadhyay,et al.  Fast and efficient Bayesian semi-parametric curve-fitting and clustering in massive data , 2012, Sankhya B.

[21]  Yee Whye Teh,et al.  Collapsed Variational Dirichlet Process Mixture Models , 2007, IJCAI.

[22]  H. J. B. Birks,et al.  An expanded calibration model for inferring lakewater and air temperatures from fossil chironomid assemblages in northern Fennoscandia , 1999 .

[23]  On Bayesian "central clustering": Application to landscape classification of Western Ghats , 2011, 1111.7105.

[24]  Kyungduk Ko,et al.  Bayesian Wavelet-Based Methods for the Detection of Multiple Changes of the Long Memory Parameter , 2006, IEEE Transactions on Signal Processing.

[25]  M. J. Bayarri,et al.  P Values for Composite Null Models , 2000 .

[26]  David Ruppert,et al.  Theory & Methods: Spatially‐adaptive Penalties for Spline Fitting , 2000 .

[27]  S. Bhattacharya.  A Fully Bayesian Approach to Assessment of Model Adequacy in Inverse Problems , 2012, 1203.2403.

[28]  S. Bhattacharya,et al.  Importance re-sampling {MCMC} for cross-validation in inverse problems , 2007 .

[29]  Sourabh Bhattacharya,et al.  A Bayesian semiparametric model for organism based environmental reconstruction , 2006 .

[30]  D. Blackwell,et al.  Ferguson Distributions Via Polya Urn Schemes , 1973 .

[31]  A. SenGupta,et al.  Bayesian analysis of semiparametric linear-circular models , 2009 .

[32]  Analabha Basu,et al.  A Novel Bayesian Semiparametric Algorithm for Inferring Population Structure and Adjusting for Case‐Control Association Tests , 2013, Biometrics.

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