IMSPE {hetGP}

Integrated Mean Square Prediction Error

Description

IMSPE of a given design

Usage

IMSPE(
  X,
  theta = NULL,
  Lambda = NULL,
  mult = NULL,
  covtype = NULL,
  nu = NULL,
  eps = sqrt(.Machine$double.eps)
)

Arguments

X

hetGP or homGP model. Alternatively, one can provide a matrix of unique designs considered

theta

lengthscales

Lambda

diagonal matrix for the noise

mult

number of replicates at each design

covtype

either "Gaussian", "Matern3_2" or "Matern5_2"

nu

variance parameter

eps

numerical nugget

Details

One can provide directly a model of class hetGP or homGP, or provide X and all other arguments