Explorer A Multi-resolution Gaussian process model for the analysis of large spatial data sets

Abstract A multi-resolution basis is developed to predict two-dimensional spatial fields based on irregularly spaced observations. The basis functions at each level of resolution are constructed as radial basis functions using a Wendland compactly supported correlation function with the nodes arranged on a rectangular grid. The grid at each finer level increases by a factor of two and the basis functions are scaled to have a constant overlap. The coefficients associated with the basis functions at each level of resolution are distributed according to a Gaussian Markov random field (GMRF) and take advantage of the fact that the basis is organized as a lattice. Several numerical examples and analytical results establish that this scheme gives a good approximation to standard covariance functions such as the Matérn and also has flexibility to fit more complicated shapes. The other important feature of this model is that it can be applied to statistical inference for large spatial datasets because key matrices in the computations are sparse. The computational efficiency applies to both the spatial predictions and also the evaluation of the likelihood. Although our framework has similarities to fixed rank Kriging, the model gives a better approximation to situations where the nugget variance is small and the spatial process is close to interpolating the observations.

[1]  M. Fuentes.  Approximate Likelihood for Large Irregularly Spaced Spatial Data , 2007, Journal of the American Statistical Association.

[2]  Jianhua Z. Huang,et al.  A full scale approximation of covariance functions for large spatial data sets , 2012 .

[3]  Richard L. Smith,et al.  Asymptotic properties of computationally efficient alternative estimators for a class of multivariate normal models , 2007 .

[4]  H. Rue,et al.  An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach , 2011 .

[5]  Holger Wendland,et al.  Error Estimates for Interpolation by Compactly Supported Radial Basis Functions of Minimal Degree , 1998 .

[6]  Peter Congdon,et al.  Gaussian Markov Random Fields: Theory and Applications , 2007 .

[7]  N. Cressie,et al.  Fixed rank kriging for very large spatial data sets , 2008 .

[8]  C. Willmott,et al.  CLIMATOLOGICALLY AIDED INTERPOLATION (CAI) OF TERRESTRIAL AIR TEMPERATURE , 1995 .

[9]  Zhiyi Chi,et al.  Approximating likelihoods for large spatial data sets , 2004 .

[10]  Christian P. Robert,et al.  Statistics for Spatio-Temporal Data , 2014 .

[11]  Michael L. Stein,et al.  A modeling approach for large spatial datasets , 2008 .

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

[13]  Stephan R. Sain,et al.  spam: A Sparse Matrix R Package with Emphasis on MCMC Methods for Gaussian Markov Random Fields , 2010 .

[14]  Holger Wendland,et al.  Piecewise polynomial, positive definite and compactly supported radial functions of minimal degree , 1995, Adv. Comput. Math..

[15]  Andrew O. Finley,et al.  Approximate Bayesian inference for large spatial datasets using predictive process models , 2012, Comput. Stat. Data Anal..

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

[17]  P. Jones,et al.  Hemispheric and large-scale land-surface air temperature variations: An extensive revision and an update to 2010: LAND-SURFACE TEMPERATURE VARIATIONS , 2012 .

[18]  David Higdon,et al.  A process-convolution approach to modelling temperatures in the North Atlantic Ocean , 1998, Environmental and Ecological Statistics.

[19]  H. Rue,et al.  Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations , 2009 .

[20]  Matthias Katzfuss,et al.  Spatio‐temporal smoothing and EM estimation for massive remote‐sensing data sets , 2011 .

[21]  Finn Lindgren,et al.  Explicit construction of GMRF approximations to generalised Matérn fields on irregular grids , 2007 .

[22]  S. R. Searle,et al.  On Deriving the Inverse of a Sum of Matrices , 1981 .