Nonparametric Bayesian density estimation on manifolds with applications to planar shapes.

Statistical analysis on landmark-based shape spaces has diverse applications in morphometrics, medical diagnostics, machine vision and other areas. These shape spaces are non-Euclidean quotient manifolds. To conduct nonparametric inferences, one may define notions of centre and spread on this manifold and work with their estimates. However, it is useful to consider full likelihood-based methods, which allow nonparametric estimation of the probability density. This article proposes a broad class of mixture models constructed using suitable kernels on a general compact metric space and then on the planar shape space in particular. Following a Bayesian approach with a nonparametric prior on the mixing distribution, conditions are obtained under which the Kullback-Leibler property holds, implying large support and weak posterior consistency. Gibbs sampling methods are developed for posterior computation, and the methods are applied to problems in density estimation and classification with shape-based predictors. Simulation studies show improved estimation performance relative to existing approaches.

[1]  I. Holopainen.  Riemannian Geometry , 1927, Nature.

[2]  L. Schwartz.  On Bayes procedures , 1965 .

[3]  G. S. Watson,et al.  Equatorial distributions on a sphere , 1965 .

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

[5]  T. Ferguson.  Prior Distributions on Spaces of Probability Measures , 1974 .

[6]  G. S. Watson.  Statistics on Spheres , 1983 .

[7]  Albert Y. Lo,et al.  On a Class of Bayesian Nonparametric Estimates: I. Density Estimates , 1984 .

[8]  D. Kendall.  SHAPE MANIFOLDS, PROCRUSTEAN METRICS, AND COMPLEX PROJECTIVE SPACES , 1984 .

[9]  J. Sethuraman.  A CONSTRUCTIVE DEFINITION OF DIRICHLET PRIORS , 1991 .

[10]  J. Kent.  The Complex Bingham Distribution and Shape Analysis , 1994 .

[11]  S. MacEachern.  Estimating normal means with a conjugate style dirichlet process prior , 1994 .

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

[13]  S. MacEachern,et al.  A semiparametric Bayesian model for randomised block designs , 1996 .

[14]  K. Mardia,et al.  Statistical Shape Analysis , 1998 .

[15]  J. Ghosh,et al.  POSTERIOR CONSISTENCY OF DIRICHLET MIXTURES IN DENSITY ESTIMATION , 1999 .

[16]  M. Escobar,et al.  Markov Chain Sampling Methods for Dirichlet Process Mixture Models , 2000 .

[17]  M. Stephens.  Dealing with label switching in mixture models , 2000 .

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

[19]  Michael,et al.  On a Class of Bayesian Nonparametric Estimates : I . Density Estimates , 2008 .

[20]  R. Bhattacharya,et al.  LARGE SAMPLE THEORY OF INTRINSIC AND EXTRINSIC SAMPLE MEANS ON MANIFOLDS—II , 2003 .

[21]  Bruno Pelletier.  Kernel density estimation on Riemannian manifolds , 2005 .

[22]  Alan S. Willsky,et al.  Nonparametric shape priors for active contour-based image segmentation , 2005, 2005 13th European Signal Processing Conference.

[23]  Stephen G. Walker,et al.  Sampling from compositional and directional distributions , 2006, Stat. Comput..

[24]  Michael A. West,et al.  Hierarchical priors and mixture models, with applications in regression and density estimation , 2006 .

[25]  Stephen G. Walker,et al.  Sampling the Dirichlet Mixture Model with Slices , 2006, Commun. Stat. Simul. Comput..

[26]  G. Roberts,et al.  Retrospective Markov chain Monte Carlo methods for Dirichlet process hierarchical models , 2007, 0710.4228.

[27]  S. MacEachern,et al.  Bayesian Density Estimation and Inference Using Mixtures , 2007 .

[28]  K. Mardia,et al.  Protein Bioinformatics and Mixtures of Bivariate von Mises Distributions for Angular Data , 2007, Biometrics.

[29]  R. Bhattacharya,et al.  Nonparametric statistics on manifolds with applications to shape spaces , 2008, 0805.3282.

[30]  S. Ghosal,et al.  Kullback Leibler property of kernel mixture priors in Bayesian density estimation , 2007, 0710.2746.

[31]  O. Papaspiliopoulos.  A note on posterior sampling from Dirichlet mixture models , 2008 .

[32]  Subhashis Ghosal,et al.  Correction to: “Kullback Leibler property of kernel mixture priors in Bayesian density estimation” Electronic J. Statist. 2 (2008) 298–331 , 2009 .

[33]  K. P. Lennox,et al.  Density Estimation for Protein Conformation Angles Using a Bivariate von Mises Distribution and Bayesian Nonparametrics , 2009, Journal of the American Statistical Association.

[34]  C. Yau,et al.  Bayesian Nonparametric Hidden Markov Models with application to the analysis of copy-number-variation in mammalian genomes. , 2011, Journal of the Royal Statistical Society. Series B, Statistical methodology.

[35]  David B. Dunson,et al.  Strong consistency of nonparametric Bayes density estimation on compact metric spaces with applications to specific manifolds , 2012, Annals of the Institute of Statistical Mathematics.