| predict {SpatialExtremes} | R Documentation |
This function predicts the marginal GEV parameters from a fitted max-stable process.
## S3 method for class 'maxstab': predict(object, newdata, ret.per = NULL, ...)
object |
An object of class ``maxstab''. Most often, it will be
the output of the function fitmaxstab. |
newdata |
An optional data frame in which to look for variables with which to predict. If omitted, the fitted values are used. |
ret.per |
Numeric vector giving the return periods for which
return levels are computed. If NULL (default), no return
levels are computed. |
... |
further arguments passed to or from other methods. |
'predict.maxstab' produces a vector of predictions or a matrix of predictions.
Mathieu Ribatet
## 1- Simulate a max-stable random field
require(RandomFields)
n.site <- 35
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")
ms0 <- MaxStableRF(locations[,1], locations[,2], grid=FALSE, model="wh",
param=c(0,1,0,3, .5), maxstable="extr",
n = 50)
## 2- Transformation to non unit Frechet margins
ms1 <- t(ms0)
param.loc <- -10 + 2 * locations[,2]
param.scale <- 5 + 2 * locations[,1]
param.shape <- rep(0.2, n.site)
for (i in 1:n.site)
ms1[,i] <- param.scale[i] * (ms1[,i]^param.shape[i] - 1) /
param.shape[i] + param.loc[i]
## 3- Fit a max-stable process with the following model for
## the GEV parameters
form.loc <- loc ~ lat
form.scale <- scale ~ lon
form.shape <- shape ~ 1
schlather <- fitmaxstab(ms1, locations, "whitmat", loc.form = form.loc,
scale.form = form.scale, shape.form =
form.shape)
## 4- GEV parameters estimates at each locations or at ungauged locations
predict(schlather)
ungauged <- data.frame(lon = runif(10, 0, 10), lat = runif(10, 0, 10))
predict(schlather, ungauged)