make_splm: Make a spatial linear model

In ericwfox/slmrf:

Make a spatial linear model

Description

Make spatial linear model object using estimated covariance parameters from optim().

Usage

make_splm(params, X, y, cov_type, dist_mtx = NULL, m2loglike = NULL,
  est_meth = NULL, hessian = NULL, dist_K = NULL, dist_S = NULL,
  knot_coord = NULL, optim_out = NULL)

Arguments

params	Numeric vector of fitted covariance parameter values (nugget, partial sill, and range, in that order).
X	Design matrix. Must include column of 1's for intercept.
y	Numeric response vector.
cov_type	character. Type of covariance model: exponential, gaussian, or spherical.
dist_mtx	Distance matrix used to compute covariance matrix.
m2loglike	Minus 2 times the log-likelihood (from optim()).
est_meth	character. Estimation method used for m2LL: 'REML' for restricted maximum likelihood, or 'ML' for maximum likelihood.
hessian	Hessian matrix from optim().
dist_K	Distance matrix for knot locations.
dist_S	Distance matric between data locations and knots (nrow is number of data points and ncol is number of knots).
knot_coord	data frame or matrix. Coordinates of knots used for reduced rank method. Optional to include (coordinates are just saved in 'splm' object for later reference).
optim_out	list output from optim().

Value

List of class splm containg features of fitted spatial linear model (e.g., covariance matrix, predictor coefficients, ...).

Author(s)

Eric W. Fox

See Also

m2LL, optim

ericwfox/slmrf documentation built on Feb. 24, 2024, 11:02 p.m.