GMRF_AR: Autoregressive GMRF

In andrewzm/atminv: Reproducible package for Section 4.1 of "Spatio-temporal bivariate statistical models for atmospheric trace-gas inversion"

Description

This function initialises an autoregressive block and represents it as a Gaussian Markov Random Field with mean mu and precision matrix Q. Only AR1 processes along the real line are implemented for now. Also a data frame can be specified with more details on the GMRF.

Usage

GMRF_AR(n = 10, a = 0.8, order = 1, precinc = 1, df = data.frame())

Arguments

n	number of vertices
a	autoregression coefficient
order	order of AR process (only AR1 implemented for now)
precinc	precision constant (multiplies the template precision matrix)
df	data frame of length n with more details (for example axis, covariate information)

Value

Object of class GMRF with zero mean

Examples

require(Matrix)
# Create a first-order random walk GMRF
my_RW <- GMRF_AR1(n=10, a = 0.8, precinc =2)
print(getPrecision(my_RW))
print(getMean(my_RW))
print(getDf(my_RW))

andrewzm/atminv documentation built on May 10, 2019, 11:14 a.m.