| RMbcw {RandomFields} | R Documentation |
RMbcw is a variogram model
that bridges between some intrinsically stationary isotropic processes
and some stationary ones. It reunifies the
RMgenfbm ‘b’, RMgencauchy ‘c’
and RMdewijsian ‘w’.
The corresponding centered semi-variogram only depends on the distance r ≥ 0 between two points and is given by
γ(r)=[(r^{α}+1)^{β/alpha}-1] / (2^{β/alpha}-1)
where 0 < α ≤ 2 and β <= 2.
RMbcw(alpha, beta, c, var, scale, Aniso, proj)
alpha |
a numerical value; should be in the interval (0,2]. |
beta |
a numerical value; should be in the interval (-infty,2]. |
c |
only for experts. If given, a not necessariy positive definite function c-γ(r) is built. |
var,scale,Aniso,proj |
optional arguments; same meaning for any
|
For betaa >0, beta<0, beta=0
we have the generalised fractal Brownian motion RMgenfbm,
the generalised Cauchy model RMgencauchy,
and the de Wisjian model RMdewijsian, respectively.
Hence its two arguments alpha and beta
allow for modelling the smoothness and a wide range of tail behaviour,
respectively.
RMbcw returns an object of class RMmodel
Martin Schlather, schlather@math.uni-mannheim.de
Schlather, M (2014) A parameteric variogram model bridging between stationary and intrinsically stationary processes. arxiv 1412.1914.
RMlsfbm is equipped with Matheron's constant c for
the fractional brownian motion,
RMgenfbm,
RMgencauchy,
RMdewijsian,
RMmodel,
RFsimulate,
RFfit.
RFoptions(seed=0) ## *ANY* simulation will have the random seed 0; set ## RFoptions(seed=NA) to make them all random again model <- RMbcw(alpha=1, beta=0.5) x <- seq(0, 10, 0.02) plot(model) plot(RFsimulate(model, x=x))