varycoef (version 0.2.9)

SVC_mle_control: Set Parameters for SVC_mle

Description

Function to set up control parameters for SVC_mle

Usage

SVC_mle_control(...)
# S3 method for default
SVC_mle_control(cov.name = c("exp", "sph"),
  tapering = NULL, cl = NULL, init = NULL, lower = NULL,
  upper = NULL, save.fitted = TRUE, profileLik = FALSE,
  mean.est = c("GLS", "OLS"), pc.prior = NULL, ...)
# S3 method for SVC_mle
SVC_mle_control(object, ...)

Arguments

...

further parameters yet to be implemented

cov.name

name of the covariance function defining the covariance matrix of the GRF. Currently, only "exp" for the exponential and "exp" for spherical covariance functions are supported.

tapering

if NULL, no tapering is applied. If a scalar is given, covariance tapering with this taper range is applied, for all GRF modelling the SVC.

cl

cluster for parallelization. Currently not supported.

init

numeric. Initial values for optimization procedure. The vector consists of p-times (alternating) scale and variance, the nugget variance and the p + p.fix mean effects

lower

lower bound for optim, default NULL sets the lower bounds to 1e-6 for covariance parameters and -Inf for mean parameters.

upper

upper bound for optim, default NULL sets the upper bounds to Inf for covariance and mean parameters.

save.fitted

logical. If TRUE, calculates the fitted values and residuals after MLE and saves them.

profileLik

logical. If TRUE, MLE is done over profile Likelihood of covariance parameters.

mean.est

if profileLik is TRUE, the means have to be estimated seperately. "GLS" uses the generalized least square estimate while "OLS" uses the ordinary least squares estiamte.

pc.prior

takes vector of \(\rho_0, \alpha_\rho, \sigma_0, \alpha_\sigma\) to compute penalized complexity priors. This regulates the optimization process. Currently, only supported for Gaussian random fields of Mat<U+00E9>rn class. Based on the idea Simpson and Fulgstad.

object

An object of class SVC_mle. The function then extracts the control settings from the particular function call used to compute object.

Value

A list with which SVC_mle can be controlled

See Also

SVC_mle

Examples

# NOT RUN {
control <- SVC_mle_control(init = rep(0.3, 10))
# or
control <- SVC_mle_control()
control$init <- rep(0.3, 10)

# }