make.scaled.dist: Scaled Euclidean distances between unique locations

Description

This fonction computes scaled Euclidean distances from whichever relevant argument it can use (see Details). The result can directly by used as input for computation of the Matérn correlation matrix. It is usually called internally by HLCor, so that users may ignore it, except if they wish to control the parametrization of the scaling through control.dist$rho.mapping.

Usage

make.scaled.dist(uniqueGeo, uniqueGeo2=NULL, distMatrix, rho, 
                 rho.mapping=seq_len(length(rho)), 
                 dist.method="Euclidean")

Arguments

uniqueGeo

A matrix of geographical coordinates (e.g. 2 columns for latitude and longitude), without replicates of the same location.

uniqueGeo2

NULL, or a second matrix of geographical coordinates, without replicates of the same location. If NULL, scaled distances among uniqueGeo locations are computed. Otherwise, scaled distances between locations in the two input matrices

distMatrix

A distance matrix.

rho

A scalar or vector of positive values. Scaled distance is computed as * rho, unless a non-trivial rho.mapping is used.

rho.mapping

A set of indices controlling which elements of the rho scale vector scales which dimension(s) of the space in which (spatial) correlation matrices of random effects are computed. Scaled distance is generally computed as

dist.method

method argument of proxy::dist function (by default, "Euclidean", but e.g. "Geodesic" can be used to compute spherical distances.

Value

A matrix or dist object. If there are two input matrices, rows of the return value correspond to rows of the first matrix.

Details

The function uses the distMatrix argument if provided, in which case rho must be a scalar. Vectorial rho (i.e., different scaling of different dimensions) is feasible only by providing uniqueGeo.

Examples

Run this code

## a biologically not very meaningful, but syntactically correct example of rho.mapping
data(blackcap)
corrHLfit(migStatus ~ 1 + Matern(1|latitude+longitude+means),data=blackcap,
          objective="p_v",HLmethod="ML",ranFix=list(nu=0.5),init.corrHLfit=list(rho=c(1,1)),
          control.dist=list(rho.mapping=c(1,1,2)))