Nonparametric Bayesian inference for the spectral density function of a random field

A powerful technique for inference concerning spatial dependence in a random field is to use spectral methods based on frequency domain analysis. Here we develop a nonparametric Bayesian approach to statistical inference for the spectral density of a random field. We construct a multi-dimensional Bernstein polynomial prior for the spectral density and devise a Markov chain Monte Carlo algorithm to simulate from the posterior of the spectral density. The posterior sampling enables us to obtain a smoothed estimate of the spectral density as well as credible bands at desired levels. Simulation shows that our proposed method is more robust than a parametric approach. For illustration, we analyse a soil data example. Copyright 2010, Oxford University Press.

[1]  Axel Tenbusch,et al.  Two-dimensional Bernstein polynomial density estimators , 1994 .

[2]  Yanbing Zheng.  Shape restriction of the multi-dimensional Bernstein prior for density functions , 2011 .

[3]  S. Ghosal,et al.  Contiguity of the Whittle measure for a Gaussian time series , 2004 .

[4]  Roger Woodard,et al.  Interpolation of Spatial Data: Some Theory for Kriging , 1999, Technometrics.

[5]  Geng-zhe Chang,et al.  An improved condition for the convexity of Bernstein-Bézier surfaces over triangles , 1984, Comput. Aided Geom. Des..

[6]  A. V. D. Vaart,et al.  Posterior convergence rates for Dirichlet mixtures of beta densities , 2008 .

[7]  V. Schmidt,et al.  Kernel Estimation of the Spectral Density of Stationary Random Closed Sets , 2004 .

[8]  J. Pitman,et al.  The two-parameter Poisson-Dirichlet distribution derived from a stable subordinator , 1997 .

[9]  Philip J. Davis,et al.  The convexity of Bernstein polynomials over triangles , 1984 .

[10]  H. Ishwaran,et al.  Markov chain Monte Carlo in approximate Dirichlet and beta two-parameter process hierarchical models , 2000 .

[11]  C. Heyde.  Smoothed periodogram asymptotics and estimation for processes and fields with possible long-range dependence , 1993 .

[12]  Mircea D. Farcas,et al.  About Bernstein polynomials , 2008 .

[13]  Noel A Cressie,et al.  Statistics for Spatial Data. , 1992 .

[14]  P. Whittle.  ON STATIONARY PROCESSES IN THE PLANE , 1954 .

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

[16]  Sonia Petrone.  On the role of mixtures in Bayesian nonparametrics , 2004 .

[17]  Sonia Petrone.  Bayesian density estimation using bernstein polynomials , 1999 .

[18]  Carol A. Gotway,et al.  Statistical Methods for Spatial Data Analysis , 2004 .

[19]  Robert Haining,et al.  Statistics for spatial data: by Noel Cressie, 1991, John Wiley & Sons, New York, 900 p., ISBN 0-471-84336-9, US $89.95 , 1993 .

[20]  Eugenio Regazzini,et al.  Approximation of Laws of Multinomial Parameters by Mixtures of Dirichlet Distributions with Applications to Bayesian Inference , 1999 .

[21]  Zhe Jiang,et al.  Spatial Statistics , 2013 .

[22]  Fernando A. Quintana,et al.  Nonparametric Bayesian data analysis , 2004 .

[23]  R. Bass,et al.  Review: P. Billingsley, Convergence of probability measures , 1971 .

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

[25]  P. Lenk.  The Logistic Normal Distribution for Bayesian, Nonparametric, Predictive Densities , 1988 .

[26]  Tomas Sauer.  Multivariate Bernstein polynomials and convexity , 1991, Comput. Aided Geom. Des..

[27]  Chao A. Hsiung,et al.  Shape restricted regression with random Bernstein polynomials , 2007, 0708.1054.

[28]  S. Ghosal,et al.  Bayesian Estimation of the Spectral Density of a Time Series , 2004 .

[29]  P. Butzer.  On Two-Dimensional Bernstein Polynomials , 1953, Canadian Journal of Mathematics.

[30]  P. Robinson.  Nonparametric spectrum estimation for spatial data , 2007 .

[31]  Sonia Petrone.  Random Bernstein Polynomials , 1999 .