make_aniso_loglik: Constructor functions for local parameter estimation.

Description Usage Arguments Value Examples

View source: R/convoSPAT_paramEst.R

Description

This function generates another function to be used within optim to obtain maximum likelihood estimates of local variance parameters lam1, lam2, eta, tausq, and sigmasq, assuming the smoothness is fixed, using a Gaussian likelihood with an anisotropic covariance structure.

Usage

1
make_aniso_loglik(locations, cov.model, data, Xmat, nugg2.var)

Arguments

locations

A matrix of locations.

cov.model

String; the covariance model.

data

A vector or matrix of data to use in the likelihood calculation.

Xmat

The design matrix for the mean model.

nugg2.var

Fixed values for the variance of the second nugget term.

Value

This function returns another function for use in optim.

Examples

1
2
3
4
## Not run: 
make_aniso_loglik( locations, cov.model, data, Xmat )

## End(Not run)

convoSPAT documentation built on May 30, 2017, 12:23 a.m.

Search within the convoSPAT package
Search all R packages, documentation and source code