Subroutine called by GeoFit.
The procedure estimates the model parameters by maximization of the
composite log-likelihood.
CompLik(copula, bivariate, coordx, coordy, coordz, coordt,
coordx_dyn, corrmodel, data, distance, flagcorr,
flagnuis, fixed, GPU, grid, likelihood, local, lower,
model, n, namescorr, namesnuis, namesparam,
numparam, numparamcorr, optimizer,
onlyvar, parallel, param,
spacetime, type, upper, varest,
weigthed, ns, X, sensitivity, MM, aniso)Returns a list from an optim call.
String; the type of copula. It can be "Clayton" or "Gaussian".
Logical; if TRUE then the data come from a bivariate random field,
otherwise from a univariate random field.
A numeric \(d \times 2\) or \(d \times 3\) matrix.
Coordinates on a sphere for a fixed radius radius
are passed in lon/lat format expressed in decimal degrees.
A numeric vector giving one dimension of
spatial coordinates; optional, default is NULL.
A numeric vector giving one dimension of
spatial coordinates; optional, default is NULL.
A numeric vector giving one dimension of temporal coordinates;
optional, default is NULL (in which case a spatial random field is assumed).
A list of \(m\) numeric \(d_t \times 2\) matrices
containing dynamic (in time) spatial coordinates; optional, default is NULL.
Numeric; the ID of the correlation model.
A numeric vector, or a \(n \times d\) matrix, or a \(d \times d \times n\) array of observations.
String; the name of the spatial distance.
Default is "Eucl" (Euclidean distance). See Details.
Numeric vector of binary values indicating which parameters of the correlation function will be estimated.
Numeric vector of binary values indicating which nuisance parameters will be estimated.
Numeric vector of parameters considered as known values.
Numeric; if NULL (default), no GPU computation is performed.
Logical; if FALSE (default), data are interpreted as vector or \(n \times d\) matrix;
if TRUE, then a \(d \times d \times n\) array is considered.
String; configuration of the composite likelihood (see GeoFit).
Numeric; number of local work-items of the GPU.
Named list; optional lower bounds for parameters when using
optimizers L-BFGS-B, nlminb, or optimize.
Names must match those in the start list.
Numeric; ID of the density associated with the likelihood objects.
Numeric; number of trials in binomial random fields.
Character vector; names of the correlation parameters.
Character vector; names of the nuisance parameters.
Character vector; names of the parameters to be maximized.
Numeric; number of parameters to be maximized.
Numeric; number of correlation parameters.
String; optimization algorithm (see optim).
Default is "Nelder-Mead". Other options: "nlm", "BFGS", "L-BFGS-B", "nlminb".
For "L-BFGS-B" and "nlminb" bounds can be provided.
For 1D optimization, optimize is used.
Logical; if TRUE (and varest is TRUE),
only the variance-covariance matrix is computed without optimizing.
Default is FALSE.
Logical; if TRUE, optimization uses optimParallel
with maximum cores, when optimizer is "L-BFGS-B".
Default is FALSE.
Numeric vector of parameter values.
Logical; if TRUE, the random field is spatio-temporal,
otherwise spatial.
String; type of likelihood object.
Default is "Pairwise" (marginal composite likelihood formed by pairwise marginal likelihoods).
Named list; optional upper bounds for parameters when using
optimizers L-BFGS-B, nlminb, or optimize.
Names must match those in the start list.
Logical; if TRUE, variance estimates and standard errors are returned.
Default is FALSE.
Logical; if TRUE, decreasing weights from a compactly supported
correlation function with compact support maxdist (maxtime) are used.
Numeric; number of (dynamic) temporal instants.
Numeric; matrix of space-time covariates in the linear mean specification.
Logical; if TRUE, the sensitivity matrix is computed.
Numeric; a non-constant fixed mean.
Logical; whether anisotropy should be considered.
Moreno Bevilacqua, moreno.bevilacqua89@gmail.com, https://sites.google.com/view/moreno-bevilacqua/home,
Víctor Morales Oñate, victor.morales@uv.cl, https://sites.google.com/site/moralesonatevictor/,
Christian Caamaño-Carrillo, chcaaman@ubiobio.cl, https://www.researchgate.net/profile/Christian-Caamano
Subroutine called by GeoFit.
The procedure estimates model parameters by maximization of the composite log-likelihood.
GeoFit