kmNuggets.init {DiceKriging}R Documentation

Fitting Kriging Models

Description

kmNuggets.init is used to give initial values to fit kriging models, in presence of noisy observations.

Usage

kmNuggets.init(model)

Arguments

model an object of class km.

Details

The procedure can be summarized in 4 stages:
1) Compute the variogram and give a first estimation of the process variance, as well as lower and upper bounds.
2) Simulate several values for the process variance, around the estimation obtained at stage 1). The number of simulations is the one given in model@control$pop.size.
3) If no initial value is provided for the other covariance parameters, simulate them uniformly inside the domain delimited by model@lower and model@upper. The number of simulations is the same as in stage 2).
4) Compute the likelihood at each simulated "point" (variance + other covariance parameters), and take the best one. This point gives the first initial values.

Value

An object of class km.

Author(s)

O. Roustant, David Ginsbourger, Ecole des Mines de St-Etienne.

See Also

km, kmNuggets


[Package DiceKriging version 1.0 Index]