Multi-Resolution Kriging Based on Markov Random Fields
Utility functions for spherical coordinate and projections.
Class "gridList"
. A description of a regular and multidimensional gr...
Icosahedral multi-resolution grids
User-friendly spatial prediction and inference using a compactly suppo...
Create a matrix with given entries on the given diagonals
Find all pairwise distances within a maximum distance.
Tailoring the LatticeKrig model to a specific geometry.
Check the LKinfo object
Summary of the LKRectangle geometry for a standard two dimensional spa...
Functions for generating a multi-resolution, compactly supported basis...
Simple function to search over covariance parameters for Lattice Krig.
Spatial prediction and inference using a compactly supported multi-res...
Functions for simulating a multi resolution process following the Latt...
Creates fixed part of spatial model.
Distance function methods for LKrigDistance
in Package LatticeKrig
...
Internal functions for LatticeKrig package.
Methods to report the locations or scales associated with the lattice ...
Miscellaneous internal functions for LatticeKrig package.
Methods and functions to support normalizing to marginal variance of o...
Method that creates the spatial autoregressive (SAR) matrix.
Create or update the LatticeKrig model object (LKinfo) for spatial fit...
Creates the alpha parameter list in LatticeKrig covariance.
Method to create a.wght component from the LKinfo
object.
Creates the lattice information for a specific geometry.
Specifying non-stationary models
Geometries to represent 2-d and 3-d spherical data.
Two dimensional radial and tensor basis functions based on a Wendland ...
Internal FORTRAN routines for working with grids and finding distances...
Method for including default information in the LKinfo object.
Data examples for the LatticeKrig Vignette
Methods for the interpolation of large spatial datasets. This package follows a "fixed rank Kriging" approach but provides a surface fitting method that can approximate standard spatial data models. Using a large number of basis functions allows for estimates that can come close to interpolating the observations (a spatial model with a small nugget variance.) Moreover, the covariance model for this method can approximate the Matern covariance family but also allows for a multi-resolution model and supports efficient computation of the profile likelihood for estimating covariance parameters. This is accomplished through compactly supported basis functions and a Markov random field model for the basis coefficients. These features lead to sparse matrices for the computations and this package makes of the R spam package for sparse linear algebra. An extension of this version over previous ones ( < 5.4 ) is the support for different geometries besides a rectangular domain. The Markov random field approach combined with a basis function representation makes the implementation of different geometries simple where only a few specific R functions need to be added with most of the computation and evaluation done by generic routines that have been tuned to be efficient. One benefit of this package's model/approach is the facility to do unconditional and conditional simulation of the field for large numbers of arbitrary points. There is also the flexibility for estimating non-stationary covariances and also the case when the observations are a linear combination (e.g. an integral) of the spatial process. Included are generic methods for prediction, standard errors for prediction, plotting of the estimated surface and conditional and unconditional simulation. See the 'LatticeKrig' GitHub repository for a vignette of this package. Development of this package was supported in part by the National Science Foundation Grant 1417857 and the National Center for Atmospheric Research.