pgs (version 0.4-0)

vol.mse.est: MSE estimation for 3D-volume predictors

Description

Compute a MSE estimate for 3D-volume predictors. The structure of interest is a random compact set. The sampling device is a uniform random lattice of figures.

Usage

vol.mse.est(fldata,mse.only=TRUE,iso=FALSE,diff2use)

Arguments

fldata

data collected for volume prediction, object of class FigLatData-class.

mse.only

TRUE (default) if only the MSE estimate must be returned.

iso

FALSE (default) if the boundary is not assumed to be isotropic.

diff2use

a family of lattice vectors defining the data covariations to be used for the MSE estimation. A matrix with as many columns as lattice vectors. Each column contains the vector coordinates in the basis defined by the lattice generating matrix. Optional: a default value is computed. If isotropy is assumed, the lattice generating matrix (3 main basis vectors). Otherwise, 3 face diagonal lattice vectors are added to the 3 main basis vectors.

Value

S.est

estimate of the boundary surface area.

deformation

used when the boundary is not assumed to be isotropic. It is assumed that there exists a volume preserving linear transformation which makes the boundary isotropic. This list component contains the estimated transformation matrix.

mse.est

MSE estimate.

Details

vol.mse.est invokes the function vol.mse.