StartParam: Initializes the Parameters for Estimation Procedures
In GeoModels: Procedures for Gaussian and Non Gaussian Geostatistical (Large) Data Analysis

StartParam

R Documentation

Initializes the Parameters for Estimation Procedures

Description

Subroutine called by the fitting procedures. The procedure initializes the parameters for the fitting procedure.

Usage

StartParam(coordx, coordy, coordz, coordt, coordx_dyn, corrmodel,
           data, distance, fcall, fixed, grid, likelihood,
           maxdist, neighb, maxtime, model, n, param,
           parscale, paramrange, radius, start, taper, tapsep,
           type, typereal, weighted, copula, X, memdist, nosym,
           p_neighb, thin_method)

Arguments

`coordx`	A numeric (`d \times 2`)-matrix or (`d \times 3`)-matrix. Coordinates on a sphere for a fixed radius `radius` are passed in lon/lat format expressed in decimal degrees.
`coordy`	A numeric vector giving 1-dimension of spatial coordinates; optional argument, default is `NULL`.
`coordz`	A numeric vector giving 1-dimension of spatial coordinates; optional argument, default is `NULL`.
`coordt`	A numeric vector assigning 1-dimension of temporal coordinates.
`coordx_dyn`	A list of `m` numeric (`d_t \times 2`)-matrices containing dynamical (in time) spatial coordinates. Optional argument, default is `NULL`.
`corrmodel`	String; the name of a correlation model.
`data`	A numeric vector or a (`n \times d`)-matrix or (`d \times d \times n`)-array of observations.
`distance`	String; the name of the spatial distance. Default is `"Eucl"` (euclidean distance). See the Section Details.
`fcall`	String; `"fitting"` to call the fitting procedure and `"simulation"` to call the simulation procedure.
`fixed`	A named list giving the values of the parameters that will be considered as known values.
`grid`	Logical; if `FALSE` (default) the data are interpreted as a vector or a (`n \times d`)-matrix; if `TRUE` then a (`d \times d \times n`)-array is considered.
`likelihood`	String; the configuration of the composite likelihood.
`maxdist`	Numeric; an optional positive value indicating the maximum spatial distance considered in the composite-likelihood computation.
`neighb`	Numeric; an optional positive integer indicating the order of neighborhood in the composite likelihood computation. See the Section Details for more information.
`maxtime`	Numeric; an optional positive value indicating the maximum temporal lag considered in the composite-likelihood computation.
`model`	String; the density associated to the likelihood objects. `"Gaussian"` is the default.
`n`	Numeric; number of trials for binomial random fields.
`param`	A numeric vector of parameter values required in the simulation procedure of random fields.
`parscale`	A numeric vector of scaling factors to improve the maximizing procedure; see `optim`.
`paramrange`	A numeric vector of parameter ranges; see `optim`.
`radius`	Numeric; the radius of the sphere in the case of lon-lat coordinates. The default is `6371`, the radius of the earth.
`start`	A named list with the initial values of the parameters that are used by the numerical routines in the maximization procedure.
`taper`	String; the name of the type of covariance matrix. It can be `"Standard"` (default) or `"Tapering"` for tapered covariance matrices.
`tapsep`	Numeric; an optional value indicating the separability parameter in the space-time adaptive taper (see Details).
`type`	String; the type of likelihood objects. Temporary value set to `"WLeastSquare"` (weighted least-square) in order to compute starting values.
`typereal`	String; the real type of likelihood objects. See `GeoFit`.
`weighted`	Logical; if `TRUE` the likelihood objects are weighted; see `GeoFit`.
`copula`	String; the type of copula.
`X`	Numeric; matrix of space-time covariates.
`memdist`	Logical; if `TRUE` then the distances in the composite likelihood are computed before the optimization.
`nosym`	Logical; if `TRUE` symmetric weights are not considered.
`p_neighb`	Numeric; a value in `(0,1]` specifying the expected fraction of nearest-neighbor pairs retained through stochastic thinning. If `1` (default), no thinning is applied and all nearest-neighbor pairs are used. If `<1`, pairs are randomly retained using independent Bernoulli sampling.
`thin_method`	String; thinning scheme used when `p_neighb<1`. Default is `"iid"` (independent Bernoulli thinning).