# variogmultiv: Function to compute multivariate empirical variogram. In spacemakeR: Spatial modelling

## Description

Compute a multivariate empirical variogram. It is strictly equivalent to summing univariate variograms.

## Usage

 `1` ```variogmultiv(Y,xy, dmin=0,dmax=max(dist(xy)),nclass=20) ```

## Arguments

 `Y` A matrix with numeric data. `xy` A matrix with coordinates of samples. `dmin` The minimum distance value at which the variogram is computed (i.e. lower bound of the first class). `dmax` The maximum distance value at which the variogram is computed (i.e. higher bound of the last class). `nclass` Number of classes of distances.

## Value

A list:

 `d ` Distances (i.e. centers of distance classes). `var ` Empirical semi-variances. `n.w ` Number of connections between samples for a given distance. `n.c ` Number of samples used for the computation of the variogram. `dclass ` Character vector with the names of the distance classes.

Stephane Dray

## References

Wagner H. H. (2003) Spatial covariance in plant communities: integrating ordination, geostatistics, and variance testing. Ecology 84, 1045–1057.

## Examples

 ```1 2 3 4 5 6``` ```data(oribatid) fau <- sqrt(oribatid\$fau/outer(apply(oribatid\$fau,1,sum),rep(1,ncol(oribatid\$fau)),"*")) # Hellinger transformation faudt <- resid(lm(as.matrix(fau)~as.matrix(oribatid\$xy))) # Removing linear effect mvspec<-variogmultiv(faudt,oribatid\$xy,nclass=20) mvspec plot(mvspec\$d,mvspec\$var,ty='b',pch=20,xlab="Distance", ylab="C(distance)") ```

spacemakeR documentation built on May 31, 2017, 2:24 a.m.