predict.kmModel: Predict a 'kmModel'
In schiffner/mobKriging: Recursive Partitioning of Kriging Models

Description Usage Arguments Value References See Also Examples

Predict a fitted Kriging model.

## S3 method for class 'kmModel'
predict(object, newdata = NULL, ...)

## S3 method for class 'kmModel'
fitted(object, ...)

`object`	[`kmModel`] The fitted Kriging model.
`newdata`	[`data.frame`] The cases to be predicted.
`...`	Additional arguments to `predict.km`.

See predict.km. A list containing at least the Kriging mean and trend computed at newdata.

Roustant, O., Ginsbourger, D. and Deville, Y. (2012), DiceKriging, DiceOptim: Two R packages for the analysis of computer experiments by Kriging-based metamodeling and optimization. Journal of Statistical Software, 51(1), http://www.jstatsoft.org/.

Predict, predict, predict.km, kmModel.

## We use the first example in the documentation of function km
if (require(DiceKriging)) {
    d <- 2L
    x <- seq(0, 1, length = 4L)
    design <- expand.grid(x1 = x, x2 = x)
    y <- apply(design, 1, branin)
    df <- data.frame(y = y, design)

    ## fitting the model
    mf <- dpp(kmModel, y ~ ., data = df)
    m <- fit(kmModel, mf, formula = ~ .)

    ## fitted values and prediction
    fitted(m, type = "UK")
    predict(m, type = "UK", se.compute = FALSE, light.return = TRUE)
}