LatticeKrig (version 9.3.0)

LKrigSetupAlpha: Creates the alpha parameter list in LatticeKrig covariance.

Description

This function is called by LKrigSetup and creates the list for the alpha parameters based on the information from the call to LKrigSetup and additional information and conditions related to the geometry.

Usage

LKrigSetupAlpha(object, ...)
# S3 method for default
LKrigSetupAlpha(object, ...)
LKFindAlphaVarianceWeights(x1, LKinfo, level)

Value

A list with nlevel components each representing the alpha values at that level. In the simplest case a vector of alpha values is converted to a list.

  LKinfo<- LKrigSetup( x, alpha=c( 1,.2,.01),
                   nlevel=3, a.wght=4.5, NC= 3)
  LKrigSetupAlpha( LKinfo)
[[1]]
[1] 1
[[2]]
[1] 0.2
[[3]]
[1] 0.01

The lower level function LKFindAlphaVarianceWeights

is used to supply weights as in the default model but also evaluate the weights at arbitrary locations. This second role is part of specifying a nonstationary model where the weights vary over the spatial domain.

Arguments

x1

Locations to evaluate the alpha weights.

LKinfo

An LKinfo object that is usually created by LKrigSetup.

level

Level of the multiresolution.

object

The partial LKinfo object created within LKrigSetup

...

Any additional arguments to this method

Author

Doug Nychka

Details

The main function of this method is to fill in the sequence of alpha values for a parametric model and convert those values to a list instead of a vector. In the case that the scalar nu is supplied it is used to create the list according to:

	
	alpha <- 2**(-2 * (1:nlevel) * nu)
    alpha <- alpha/sum(alpha)
    as.list( alpha)

See Also

LKrigSetup, LKrigSetupAwght, LKrigSAR, LKrig

Examples

# an x that is just the limits of the domain	
  x<- cbind( c(0,1), c(0,1))
  
  LKinfo<- LKrigSetup( x, alpha=c( 1,.2,.01),
                   nlevel=3, a.wght=4.5, NC= 3)
  alphaList<- LKrigSetupAlpha( LKinfo)

  LKinfo<- LKrigSetup( x, nu=1, nlevel=4, a.wght=4.5, NC= 4)
  alphaList<- LKrigSetupAlpha( LKinfo)