variograms {compositions} R Documentation

## Variogram functions

### Description

Valid scalar variogram model functions.

### Usage

vgram.sph( h , nugget = 0, sill = 1, range= 1,... )
vgram.exp( h , nugget = 0, sill = 1, range= 1,... )
vgram.gauss( h , nugget = 0, sill = 1, range= 1,... )
vgram.cardsin( h , nugget = 0, sill = 1, range= 1,... )
vgram.lin( h , nugget = 0, sill = 1, range= 1,... )
vgram.pow( h , nugget = 0, sill = 1, range= 1,... )
vgram.nugget( h , nugget = 1,...,tol=1E-8 )


### Arguments

 h a vector providing distances, a matrix of distance vectors in its rows or a data.frame of distance vectors. nugget The size of the nugget effect (i.e. the limit to 0). At zero itself the value is always 0. sill The sill (i.e. the limit to infinity) range The range parameter. I.e. the distance in which sill is reached or if this does not exist, where the value is in some sense nearly the sill. ... not used tol The distance that is considered as nonzero.

### Details

The univariate variograms are used in the CompLinCoReg as building blocks of multivariate variogram models.

sph

Spherical variogram

exp

Exponential variogram

gauss

The Gaussian variogram.

gauss

The cardinal sine variogram.

lin

Linear Variogram. Increases over the sill, which is reached at range.

pow

The power variogram. Increases over the sill, which is reached at range.

nugget

The pure nugget effect variogram.

### Value

A vector of size NROW(h), giving the variogram values.

### Author(s)

K.Gerald v.d. Boogaart http://www.stat.boogaart.de

### References

Cressie, N.C. (1993) Spatial statistics

Tolosana, van den Boogaart, Pawlowsky-Glahn (2009) Estimating and modeling variograms of compositional data with occasional missing variables in R, StatGis09

vgram2lrvgram, CompLinModCoReg, vgmFit

### Examples

## Not run:
data(juraset)
X <- with(juraset,cbind(X,Y))
comp <- acomp(juraset,c("Cd","Cu","Pb","Co","Cr"))
lrv <- logratioVariogram(comp,X,maxdist=1,nbins=10)
plot(lrv)

## End(Not run)


[Package compositions version 2.0-8 Index]