| RMS {RandomFields} | R Documentation |
RMS is an operator that modifies the variance and the
coordinates or distances of a submodel φ by
C(h) = v * φ(A*h/s).
Most users will never call RMS directly, see the
details.
RMS(phi, var, scale, Aniso, proj, anisoT)
phi |
submodel |
var |
is the optional variance parameter v, It can be also an arbitrary non-negative function. |
scale |
scaling parameter s which is positive |
Aniso |
matrix or |
proj |
is the optional projection vector which defines a diagonal
matrix of zeros and ones and |
anisoT |
the transpose of the anisotropy matrix B, multiplied from the left by a distance vector x, i.e. x^\top B. |
The call in the usage section is equivalent to
phi(..., var, scale, anisoT, Aniso, proj), where phi has
to be replaced by a valid RMmodel
Most users will never call RMS directly.
RMS returns an object of class RMmodel.
At most one of the arguments,
Aniso, anisoT and proj may be given at the same time.
Martin Schlather, schlather@math.uni-mannheim.de
RMprod for an alternative way to define
an arbitrary, location dependent variance. There the standard
deviation is given so that RMprod might be used
even in the multivariate case.
RFoptions(seed=0) ## *ANY* simulation will have the random seed 0; set ## RFoptions(seed=NA) to make them all random again model1 <- RMS(RMexp(), scale=2) model2 <- RMexp(scale=2) x <- seq(0, 10, 0.02) print(all(RFcov(model1, x) == RFcov(model2, x))) # TRUE