DSpars {gmGeostats}

Create a parameter set specifying a direct sampling algorithm

Description

Create a parameter set describing a direct sampling algorithm to multipoint simulation. All parameters except nsim are optional, as they have default values reasonable according to experience.

Usage

DSpars(
  nsim = 1,
  scanFraction = 0.25,
  patternSize = 10,
  gof = 0.05,
  seed = NULL,
  ...
)

Arguments

nsim

number of realisations desired (attention: current algorithm is slow, start with small values!)

scanFraction

maximum fraction of the training image to be scanned on each iteration

patternSize

number of observations used for conditioning the simulation

gof

maximum acceptance discrepance between a data event in the training image and the conditioning data event

seed

an object specifying if and how the random number generator should be initialized, see ?simulate in base "stats" package

...

further parameters, not used

Value

an S3-list of class "gmDirectSamplingParameters" containing the six elements given as arguments to the function. This is just a compact way to provide further functions such as predict_gmSpatialModel with appropriate triggers for choosing a prediction method or another, in this case for triggering direct sampling.

Examples

(dsp = DSpars(nsim=100, scanFraction=75, patternSize=6, gof=0.05))
## then run predict(..., pars=dsp)