Page 1 of 1

[SOLVED] Average of several experimental variograms

PostPosted: Thu May 10, 2018 3:49 pm
by Marise
Dear Sir,

I have used RGeostats to define a bivariate Gaussian field, generate realizations and calculate experimental variograms, as follows.
--------------------------------------------------------------------------------------

library(Rcpp)
library('RGeostats')

N=50
db=db.create(flag.grid=T,nx=c(N,N),x0=c(0,0),ndim=2)
modY1=model.create(vartype="Gaussian",range=c(10),ndim=2,sill=1)
db=simtub(dbout=db,model=modY1,seed=1,nbsimu=10,nbtuba=2000)

vario_exp1a=vario.calc(db,nlag=30,toldis=0.5,tolang=45,calcul="vg",
means=T)

var_exp1=vario.reduce(vario_exp1a,vars=1)
var_exp2=vario.reduce(vario_exp1a,vars=2)
#and so on for the other eight variograms

--------------------------------------------------------------------------------------

Now, I would like to calculate the averages of the experimental variograms.
I have a figure in mind that resembles Fig. 3 in the paper 'A pairwise likelihood approach for the empirical estimation of the underlying variograms in the plurigaussian models' by N. Desassis et al.: https://arxiv.org/abs/1510.02668.
These figures display the averages of the variograms, as well as the 5th, 25th, 75th and 95th percentiles of the models.
Would you be able to explain to me how such figures are generated with RGeostats?

I have tried:
var_exp1$vardirs,
which prints all information about var_exp1. The slot 'gg' contains the information I want (the values at the different lag distances), but I cannot isolate it.

Many thanks indeed,

Marise Westbroek

Re: Average of several experimental variograms

PostPosted: Sat Jun 30, 2018 10:18 pm
by Didier Renard
This function has been coded (since Release V11.2.3: vario.average