f_mc_kernels: Calculate mixture component kernel matrices.

In markdrisser/convoSPAT: Convolution-Based Nonstationary Spatial Modeling

Description

f_mc_kernels calculates spatially-varying mixture component kernels using generalized linear models for each of the eigenvalues (lam1 and lam2) and the angle of rotation (eta).

Usage

f_mc_kernels(y.min = 0, y.max = 5, x.min = 0, x.max = 5, N.mc = 3^2,
  lam1.coef = c(-1.3, 0.5, -0.6), lam2.coef = c(-1.4, -0.1, 0.2),
  logit.eta.coef = c(0, -0.15, 0.15))

Arguments

y.min	Lower bound for the y-coordinate axis.
y.max	Upper bound for the y-coordinate axis.
x.min	Lower bound for the y-coordinate axis.
x.max	Upper bound for the y-coordinate axis.
N.mc	Number of mixture component locations.
lam1.coef	Log-linear regression coefficients for lam1; the coefficients correspond to the intercept, longitude, and latitude.
lam2.coef	Log-linear regression coefficients for lam2; the coefficients correspond to the intercept, longitude, and latitude.
logit.eta.coef	Scaled logit regression coefficients for eta; the coefficients correspond to the intercept, longitude, and latitude.

Value

A list with the following components:

mc.locations	A N.mc x 2 matrix of the mixture component locations.
mc.kernels	A N.mc x 2 x 2 array of kernel matrices corresponding to each of the mixture component locations.

Examples

f_mc_kernels( y.min = 0, y.max = 5, x.min = 0,
x.max = 5, N.mc = 3^2, lam1.coef = c(-1.3, 0.5, -0.6),
lam2.coef = c(-1.4, -0.1, 0.2), logit.eta.coef = c(0, -0.15, 0.15) )

markdrisser/convoSPAT documentation built on May 21, 2019, 11:48 a.m.