Description Usage Format Value References See Also Examples
An object of class StatModel
that provides infra-structure for an unfitted Kriging model.
1 |
1 2 3 4 5 6 7 8 9 10 | Formal class 'StatModel' [package "modeltools"] with 5 slots
..@ name : chr "kriging model"
..@ dpp :function (formula, data = list(), subset = NULL, na.action = NULL, frame = NULL,
enclos = sys.frame(sys.nframe()), other = list(), designMatrix = TRUE,
responseMatrix = TRUE, setHook = NULL, ...)
..@ fit :function (object, weights = NULL, noise.var = NULL, km.args = NULL, ...)
..@ predict :function (object, newdata = NULL, ...)
..@ capabilities:Formal class 'StatModelCapabilities' [package "modeltools"] with 2 slots
.. .. ..@ weights: logi FALSE
.. .. ..@ subset : logi FALSE
|
Slot fit
returns an object of class kmModel
.
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/.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 | ## 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 using kmModel:
# data pre-processing
mf <- dpp(kmModel, y ~ ., data = df)
# no trend (formula = ~ 1)
m1 <- fit(kmModel, mf)
# linear trend (formula = ~ x1 + x2)
m1 <- fit(kmModel, mf, formula = ~ .)
# predictions on the training data
# recommended: improved version of predict for models fitted with objects
# of class StatModel
Predict(m1, type = "UK")
# also possible
predict(m1, type = "UK")
## This is equivalent to:
# no trend (formula = ~ 1)
m2 <- km(design = design, response = y)
# linear trend (formula = ~ x1 + x2)
m2 <- km(formula = ~ ., design = design, response = y)
# predictions on the training data
predict(m2, newdata = design, type = "UK")
## extract information
coef(m1)
residuals(m1)
logLik(m1)
## diagnostic plots
plot(m1)
}
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.