Package 'SpatialExtremes'

Title: Modelling Spatial Extremes
Description: Tools for the statistical modelling of spatial extremes using max-stable processes, copula or Bayesian hierarchical models. More precisely, this package allows (conditional) simulations from various parametric max-stable models, analysis of the extremal spatial dependence, the fitting of such processes using composite likelihoods or least square (simple max-stable processes only), model checking and selection and prediction. Other approaches (although not completely in agreement with the extreme value theory) are available such as the use of (spatial) copula and Bayesian hierarchical models assuming the so-called conditional assumptions. The latter approaches is handled through an (efficient) Gibbs sampler. Some key references: Davison et al. (2012) <doi:10.1214/11-STS376>, Padoan et al. (2010) <doi:10.1198/jasa.2009.tm08577>, Dombry et al. (2013) <doi:10.1093/biomet/ass067>.
Authors: Mathieu Ribatet [aut, cre], Richard Singleton [ctb], R Core team [ctb]
Maintainer: Mathieu Ribatet <[email protected]>
License: GPL (>= 2)
Version: 2.1-0
Built: 2024-11-02 06:23:09 UTC
Source: CRAN

Help Index


Anova Tables

Description

Computes analysis of deviance for objects of class 'maxstab' or 'spatgev'.

Usage

## S3 method for class 'maxstab'
anova(object, object2, method = "RJ", square = "chol", ...)
## S3 method for class 'spatgev'
anova(object, object2, method = "RJ", square = "chol", ...)

Arguments

object, object2

Two objects of class 'maxstab' or 'spatgev'.

method

Character string. Must be one of "CB" or "RJ" for the Chandler and Bate or the Rotnitzky and Jewell approaches respectively. See function profile.

square

The choice for the matrix square root. This is only useful for the 'CB' method. Must be one of 'chol' (Cholesky) or 'svd' (Singular Value Decomposition).

...

Other options to be passed to the anova function.

Details

As ”maxstab” objects are fitted using pairwise likelihood, the model is misspecified. As a consequence, the likelihood ratio statistic is no longer χ2\chi^2 distributed. To compute the anova table, we use the methodology proposed by Rotnitzky and Jewell to adjust the distribution of the likelihood ratio statistic.

Value

This function returns an object of class anova. These objects represent analysis-of-deviance tables.

Author(s)

Mathieu Ribatet

References

Chandler, R. E. and Bate, S. (2007) Inference for clustered data using the independence loglikelihood Biometrika, 94, 167–183.

Rotnitzky, A. and Jewell, N. (1990) Hypothesis testing of regression parameters in semiparametric generalized linear models for cluster correlated data. Biometrika 77, 485–497.

See Also

fitmaxstab, fitspatgev, profile, TIC

Examples

##Define the coordinates of each location
n.site <- 30
locations <- matrix(rnorm(2*n.site, sd = sqrt(.2)), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(50, locations, cov.mod = "gauss", cov11 = 100, cov12 =
25, cov22 = 220)

##Now define the spatial model for the GEV parameters
param.loc <- -10 + 2 * locations[,2]
param.scale <- 5 + 2 * locations[,1] + locations[,2]^2
param.shape <- rep(0.2, n.site)

##Transform the unit Frechet margins to GEV 
for (i in 1:n.site)
  data[,i] <- frech2gev(data[,i], param.loc[i], param.scale[i],
param.shape[i])

##Define three models for the GEV margins to be fitted
loc.form <- loc ~ lat
scale.form <- scale ~ lon + I(lat^2)
shape.form <- shape ~ lon

M0 <- fitspatgev(data, locations, loc.form, scale.form, shape.form)
M1 <- fitspatgev(data, locations, loc.form, scale.form, shape.form,
shapeCoeff2 = 0)

##Model selection
anova(M0, M1)
anova(M0, M1, method = "CB", square = "svd")

Pairwise empirical and extremal concurrence probabilities

Description

This function computes the pairwise empirical or the pairwise extremal concurrence probability estimates.

Usage

concprob(data, coord, fitted, n.bins, add = FALSE, xlim = c(0,
max(dist)), ylim = c(min(0, concProb), max(1, concProb)), col = 1:2,
which = "kendall", xlab, ylab, block.size = floor(nrow(data)^(1/3)),
plot = TRUE, compute.std.err = FALSE, ...)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

fitted

An object of class maxstab - usually the output of the fitmaxstab function. May be missing.

n.bins

The number of bins to be used. If missing, pairwise F-madogram estimates will be computed.

xlim, ylim

A numeric vector of length 2 specifying the x/y coordinate ranges.

col

The colors used for the points and optionnaly the fitted curve.

which

A character string specifying which estimator should be used. Should be one of "emp" (empirical), "boot" (bootstrap version) and "kendall" (kendall based).

xlab, ylab

The labels for the x/y-axis (may be missing).

add

Logical. If TRUE, the plot is added to the current figure; otherwhise (default) a new plot is computed.

block.size

Integer specifying the block size for the empirical and bootstrap estimator.

plot

Logical. If TRUE (default) a plot is produced.

compute.std.err

Logical. If TRUE, standard errors are estimated using a jackknife procedure. It is currently only available for the Kendall estimator.

...

Additional options to be passed to the plot function.

Value

This function returns invisibly a matrix containing the pairwise distances and the concurrence probability estimates.

Author(s)

Mathieu Ribatet

References

Dombry, C., Ribatet, M. and Stoev, S. (2017) Probabilities of concurrent extremes. To appear in JASA

See Also

fmadogram, lmadogram

Examples

n.site <- 25
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
n.obs <- 100
data <- rmaxstab(n.obs, locations, cov.mod = "whitmat", nugget = 0, range = 1,
smooth = 1.75)

##Compute the F-madogram
concprob(data, locations)

##Compare the F-madogram with a fitted max-stable process
fitted <- fitmaxstab(data, locations, "whitmat", nugget = 0)
concprob(fitted = fitted)

Maps of concurrence probabilities/expected concurrence cell area

Description

This function produces maps for concurrence probabilities or expected concurrence cell areas.

Usage

concurrencemap(data, coord, which = "kendall", type = "cell", n.grid =
100, col = cm.colors(64), plot = TRUE, plot.border = NULL,
compute.std.err = FALSE, ...)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

which

A character string specifying which estimator should be used. Should be one of "emp" (empirical), "boot" (bootstrap version) and "kendall" (kendall based).

type

Either "cell" for cell areas or a integer between 1 and the number of locations specifying which site should be used as reference location—see Details.

n.grid

Integer specifying the size of the prediction grid.

col

The colors used to produce the map.

plot

Logical. If TRUE (default), a map is produced; otherwise results are invisibly returned.

plot.border

The name of an R function that can be used to plot the border of the study region. If NULL, no border are plotted.

compute.std.err

Logical. If TRUE, a map of standard errors is also produced. It is currently only available for concurrence probability maps.

...

Additional options to be passed to the image function.

Value

This function returns invisibly a list with the x and y coordinates and the corresponding values for the estimated concurrence probabilities or expected concurrence cell area.

Author(s)

Mathieu Ribatet

References

Dombry, C., Ribatet, M. and Stoev, S. (2015) Probabilities of concurrent extremes. Submitted

See Also

concprob

Examples

##require(maps) ## <<-- to plot US borders
data(USHCNTemp)
coord <- as.matrix(metadata[,2:3])

## Subset the station to have a fast example
n.site <- 30
chosen.site <- sample(nrow(coord), n.site)
coord <- coord[chosen.site,]
maxima.summer <- maxima.summer[,chosen.site]

## Define a function to plot the border
border <- function(add = FALSE) maps::map("usa", add = add)

par(mar = rep(0, 4))
## Produce a pairwise concurrence probability map w.r.t. station number 15
concurrencemap(maxima.summer, coord, type = 15, plot.border = border,
 compute.std.err = TRUE)

## Produce the expected concurrence cell area
concurrencemap(maxima.summer, coord, plot.border = border)

Produces a conditional 2D map from a fitted max-stable process

Description

Produces a conditional 2D map from a fitted max-stable process.

Usage

condmap(fitted, fix.coord, x, y, covariates = NULL, ret.per1 = 100,
ret.per2 = ret.per1, col = terrain.colors(64), plot.contour = TRUE,
...)

Arguments

fitted

An object of class maxstab. Most often, it will be the output of the function fitmaxstab.

fix.coord

The spatial coordinates of the location from which the conditional quantile is computed.

x, y

Numeric vector defining the grid at which the levels are computed.

covariates

An array specifying the covariates at each grid point defined by x and y. If NULL, no covariate is needed. See map to see how to build it.

ret.per1, ret.per2

Numerics giving the return period for which the quantile map is plotted. See details.

col

A list of colors such as that generated by 'rainbow', 'heat.colors', 'topo.colors', 'terrain.colors' or similar functions.

plot.contour

Logical. If TRUE (default), contour lines are added to the plot.

...

Several arguments to be passed to the image function.

Details

The function solves the following equation:

Pr[Z(x2)>z2Z(x1)>z1]=1T2\Pr\left[Z(x_2) > z_2 | Z(x_1) > z_1 \right] = \frac{1}{T_2}

where z1=1/log(11/T1)z_1 = -1 / \log(1 - 1/T_1).

In other words, it computes, given that at location x1x_1 we exceed the level z1z_1, the levels which is expected to be exceeded in average every T2T_2 year.

Value

A plot. Additionally, a list with the details for plotting the map is returned invisibly.

Author(s)

Mathieu Ribatet

See Also

map, filled.contour, heatmap, heat.colors, topo.colors, terrain.colors, rainbow

Examples

##Define the coordinate of each location
n.site <- 30
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(50, locations, cov.mod = "whitmat", nugget = 0, range =
2, smooth = 1)

##Now define the spatial model for the GEV parameters
param.loc <- -10 - 4 * locations[,1] + locations[,2]^2
param.scale <- 5 + locations[,2] + locations[,1]^2 / 10
param.shape <- rep(.2, n.site)

##Transform the unit Frechet margins to GEV
for (i in 1:n.site)
  data[,i] <- frech2gev(data[,i], param.loc[i], param.scale[i],
param.shape[i])

##Define a model for the GEV margins to be fitted
##shape ~ 1 stands for the GEV shape parameter is constant
##over the region
loc.form <- loc ~ lon + I(lat^2)
scale.form <- scale ~ lat + I(lon^2)
shape.form <- shape ~ 1

##  1- Fit a max-stable process
fitted <- fitmaxstab(data, locations, "whitmat", loc.form, scale.form,
                     shape.form, nugget = 0)

cond.coord <- c(5.1, 5.1)
condmap(fitted, cond.coord, seq(0, 10, length = 25), seq(0,10, length
 =25), ret.per1 = 100, ret.per2 = 1.5)
points(t(cond.coord), pch = "*", col = 2, cex = 2)

Conditional simulation of Gaussian random fields

Description

This function generates conditional simulation of Gaussian random fields from the simple kriging predictor.

Usage

condrgp(n, coord, data.coord, data, cov.mod = "powexp", mean = 0, sill =
1, range = 1, smooth = 1, grid = FALSE, control = list())

Arguments

n

Integer. The number of conditional simulations.

coord

A numeric vector or matrix specifying the coordinates where the process has to be generated. If coord is a matrix, each row specifies one location.

data.coord

A numeric vector or matrix specifying the coordinates where the process is conditioned.

data

A numeric vector giving the conditioning observations.

cov.mod

A character string specifying the covariance function family. Must be one of "whitmat", "powexp", "cauchy" or "bessel" for the Whittle-Mater, the powered exponential, the Cauchy or Bessel covariance families.

mean, sill, range, smooth

The mean, sill, range and smooth of the Gaussian process.

grid

Logical. Does coord specifies a grid?

control

A named list passing options to the simulation method of Gaussian processes — see rgp.

Value

A list with components:

coord

The coordinates at which the process was simulated;

cond.sim

The simulated process;

data.coord

The coordinates of the conditioning locations;

data

The conditioning observations;

cov.mod

The covariance function family;

grid

Does coord specifies a grid?

Author(s)

Mathieu Ribatet

See Also

kriging, rgp.

Examples

## Several conditional simulations
n.site <- 50
n.sim <- 512

x.obs <- runif(n.site, -100, 100)
x.sim <- seq(-100, 100, length = n.sim)

data <- rgp(1, x.obs, "whitmat", sill = 1, range = 10, smooth = 0.75)

sim <- condrgp(5, x.sim, x.obs, data, "whitmat", sill = 1, range =
10, smooth = 0.75)

matplot(x.sim, t(sim$cond.sim),  type = "l", lty = 1, xlab = "x", ylab =
expression(Y[cond](x)))
points(x.obs, data, pch = 21, bg = 1)
title("Five conditional simulations")

## Comparison between one conditional simulations and the kriging
## predictor on a grid
x.obs <- matrix(runif(2 * n.site, -100, 100), ncol = 2)
x <- y <- seq(-100, 100, length = 100)
x.sim <- cbind(x, y)

data <- rgp(1, x.obs, "whitmat", sill = 1, range = 50, smooth = 0.75)

krig <- kriging(data, x.obs, x.sim, "whitmat", sill = 1, range = 50,
smooth = 0.75, grid = TRUE)
sim <- condrgp(1, x.sim, x.obs, data, "whitmat", sill = 1, range = 50,
smooth = 0.75, grid = TRUE)

z.lim <- range(c(sim$cond.sim, data, krig$krig.est))
breaks <- seq(z.lim[1], z.lim[2], length = 65)
col <- heat.colors(64)
idx <- as.numeric(cut(data, breaks))

op <- par(mfrow = c(1,2))
image(x, y, krig$krig.est, col = col, breaks = breaks)
points(x.obs, bg = col[idx], pch = 21)
title("Kriging predictor")
image(x, y, sim$cond.sim, col = col, breaks = breaks)
points(x.obs, bg = col[idx], pch = 21)
title("Conditional simulation")
## Note how the background colors of the above points matches the ones
## returned by the image function
par(op)

Conditional simulation of max-linear random fields

Description

This function generates (approximate) conditional simulation of unit Frechet max-linear random fields. It can be used to get approximate conditional simulation for max-stable processes.

Usage

condrmaxlin(n, coord, data.coord, data, cov.mod = "gauss", ..., grid =
FALSE, p = 10000)

Arguments

n

Integer. The number of conditional simulations.

coord

A numeric vector or matrix specifying the coordinates where the process has to be generated. If coord is a matrix, each row specifies one locations.

data.coord

A numeric vector or matrix specifying the coordinates where the process is conditioned.

data

A numeric vector giving the conditioning observations.

cov.mod

A character string specifying the max-stable model. See section Details.

...

The parameters of the max-stable model. See section Details.

grid

Logical. Does coord defines a grid?

p

An integer. The number of unit Frechet random variables used in the max-linear approximation.

Details

Any unit Frechet max-stable processes {Z(x)}\{Z(x)\} can be approximated by a unit Frechet max-linear process, i.e.,

Z(x)maxj=1,,pfj(x)Zj,Z(x) \approx \max_{j=1, \ldots, p} f_j(x) Z_j,

where fjf_j are non-negative deterministic functions, pp is a sufficiently large integer and ZjZ_j are independent unit Frechet random variables. Note that to ensure unit Frechet margins, the following condition has to be satisfied

j=1,,pfj(x)=1,\sum_{j=1, \ldots, p} f_j(x) = 1,

for all xx.

Currently only the discretized Smith model is implemented for which fj(x)=c(p)φ(xuj;Σ)f_j(x) = c(p) \varphi(x - u_j ; \Sigma) where φ(;Σ)\varphi(\cdot; \Sigma) is the zero mean (multivariate) normal density with covariance matrix Σ\Sigma, uju_j is a sequence of deterministic points appropriately chosen and c(p)c(p) is a constant ensuring unit Frechet margins.

Value

A matrix containing observations from the required max-stable model. Each column represents one stations. If grid = TRUE, the function returns an array of dimension nrow(coord) x nrow(coord) x n.

Warnings

It may happen that some conditional observations are not honored because the approximation of a max-stable process by a max-linear one isn't accurate enough! Sometimes taking a larger p solves the issue.

Author(s)

Mathieu Ribatet

References

Wang, Y. and Stoev, S. A. (2011) Conditional Sampling for Max-Stable Random Fields. Advances in Applied Probability.

See Also

rmaxstab, condrmaxlin

Examples

## One dimensional conditional simulations
n.cond.site <- 10
cond.coord <- runif(n.cond.site, -10, 10)
data <- rmaxlin(1, cond.coord, var = 3, p = 10000)

x <- seq(-10, 10, length = 250)
cond.sim <- condrmaxlin(5, x, cond.coord, data, var = 3)

matplot(x, t(log(cond.sim)), type = "l", lty = 1, pch = 1)
points(cond.coord, log(data))

## Two dimensional conditional simulation
cond.coord <- matrix(runif(2 * n.cond.site, -10, 10), ncol = 2)
data <- rmaxstab(1, cond.coord, "gauss", cov11 = 4, cov12 = 0, cov22 = 4)

x <- y <- seq(-10, 10, length = 75)
cond.sim <- condrmaxlin(4, cbind(x, y), cond.coord, data, cov11 = 4,
cov12 = 0, cov22 = 4, grid = TRUE, p = 2000)
## Note p is set to 2000 for CPU reasons but is likely to be too small

op <- par(mfrow = c(2, 2), mar = rep(1, 4))
for (i in 1:4){
image(x, y, log(cond.sim[,,i]), col = heat.colors(64), xaxt = "n", yaxt
= "n", bty = "n")
contour(x, y, log(cond.sim[,,i]), add = TRUE)
text(cond.coord[,1], cond.coord[,2], round(log(data), 2), col = 3)
}
par(op)

Conditional simulation of max-stable processes

Description

This function performs conditional simulation of various max-stable processes.

Usage

condrmaxstab(k = 1, coord, cond.coord, cond.data, cov.mod = "powexp",
..., do.sim = TRUE, thin = n.cond, burnin = 50, parts)

Arguments

k

An integer. The number of conditional simulations to be generated.

coord

A vector or matrix that gives the coordinates of each location. Each row corresponds to one location - if any.

cond.coord

A vector or matrix that gives the coordinates of each conditional location. Each row corresponds to one location - if any.

cond.data

A vector that gives the conditional values at the corresponding conditioning locations. Each row corresponds to one location - if any.

cov.mod

A character string that gives the max-stable model. This must be one of "brown" for the Brown-Resnick model, or "whitmat", "cauchy", "powexp" and "bessel" for the Schlather model with the given correlation family.

...

The parameters of the max-stable model. See rmaxstab for more details.

do.sim

A logical value. If TRUE (the default), the conditional simulations are performed; otherwise only the simulated random partitions, i.e., the hitting scenarios, are returned.

thin

A positive integer giving by which amount the generated Markov chain should be thinned. This is only useful when the number of conditioning locations is greater than 7.

burnin

A positive integer giving the duration of the burnin period of the Markov chain.

parts

A matrix giving the hitting scenarios. Each row corresponds to one hitting scenarios. If missing then a Gibbs sampler will be used to generate such hitting scenarios.

Details

The algorithm consists in three steps:

  1. Draw a random partition θ\theta from

    Pr{θ=τZ(x)=z}\Pr\{\theta = \tau \mid Z(x) = z\}

  2. Given the random partition, draw the extremal functions from

    Pr{φ+Z(x)=z,θ=τ}\Pr\{\varphi^+ \in \cdot \mid Z(x) = z, \theta = \tau\}

  3. Independently, draw the sub-extremal functions, i.e.,

    maxi1φi1{φi(x)<z}\max_{i \ge 1} \varphi_i 1_{\{\varphi_i(x) < z\}}

The distribution in Step 1 is usually intractable and in such cases a random scan Gibbs sampler will be used to sample from this distribution.

Value

This function returns a list whose components are

sim

The conditional simulations. Beware the first values corresponds to the conditioning values.

sub.ext.fct

The values of the sub-extremal functions.

ext.fct

The values of the extremal functions.

timings

The timings in seconds for each step of the algorithm.

Warning

This function can be extremely time consuming when the number of conditioning locations is large.

Author(s)

Mathieu Ribatet

References

Dombry, C. and Eyi-Minko, F. (2012) Regular conditional distributions of max infinitely divisible processes. Submitted.

Dombry, C., Eyi-Minko, F. and Ribatet, M. (2012) Conditional simulation of max-stable processes. To appear in Biometrika.

See Also

rmaxstab, condrgp

Examples

n.sim <- 50
n.cond <- 5

range <- 10
smooth <- 1.5

n.site <- 200
coord <- seq(-5, 5, length = n.site)
cond.coord <- seq(-4, 4, length = n.cond)
all.coord <- c(cond.coord, coord)

all.cond.data <- rmaxstab(1, all.coord, "powexp", nugget = 0, range = range,
                      smooth = smooth)
cond.data <- all.cond.data[1:n.cond]

ans <- condrmaxstab(n.sim, coord, cond.coord, cond.data, range = range,
                    smooth = smooth, cov.mod = "powexp")

idx <- order(all.coord)
matplot(coord, t(log(ans$sim)), type = "l", col = "grey", lty = 1,
        xlab = expression(x), ylab = expression(Z(x)))
lines(all.coord[idx], log(all.cond.data)[idx])
points(cond.coord, log(cond.data), pch = 15, col = 2)

Defines and computes covariance functions

Description

This function defines and computes several covariance function either from a fitted “max-stable” model; either by specifying directly the covariance parameters.

Usage

covariance(fitted, nugget, sill, range, smooth, smooth2 = NULL, cov.mod =
"whitmat", plot = TRUE, dist, xlab, ylab, col = 1, ...)

Arguments

fitted

An object of class “maxstab”. Most often this will be the output of the fitmaxstab function. May be missing if sill, range, smooth and cov.mod are given.

nugget, sill, range, smooth, smooth2

The nugget, sill, scale and smooth parameters for the covariance function. May be missing if fitted is given.

cov.mod

Character string. The name of the covariance model. Must be one of "whitmat", "cauchy", "powexp", "bessel" or "caugen" for the Whittle-Matern, Cauchy, Powered Exponential, Bessel and Generalized Cauchy models. May be missing if fitted is given.

plot

Logical. If TRUE (default) the covariance function is plotted.

dist

A numeric vector corresponding to the distance at which the covariance function should be evaluated. May be missing.

xlab, ylab

The x-axis and y-axis labels. May be missing.

col

The color to be used for the plot.

...

Several option to be passed to the plot function.

Details

Currently, four covariance functions are defined: the Whittle-Matern, powered exponential (also known as "stable"), Cauchy and Bessel models. These covariance functions are defined as follows for h>0h > 0

Whittle-Matern

γ(h)=σ21κΓ(κ)(hλ)κKκ(hλ)\gamma(h) = \sigma \frac{2^{1-\kappa}}{\Gamma(\kappa)} \left(\frac{h}{\lambda} \right)^{\kappa} K_{\kappa}\left(\frac{h}{\lambda} \right)

Powered Exponential

γ(h)=σexp[(hλ)κ]\gamma(h) = \sigma \exp \left[- \left(\frac{h}{\lambda} \right)^{\kappa} \right]

Cauchy

γ(h)=σ[1+(hλ)2]κ\gamma(h) = \sigma \left[1 + \left(\frac{h}{\lambda} \right)^2 \right]^{-\kappa}

Bessel

γ(h)=σ(2λh)κΓ(κ+1)Jκ(hλ)\gamma(h) = \sigma \left(\frac{2 \lambda}{h}\right)^{\kappa} \Gamma(\kappa + 1) J_{\kappa}\left(\frac{h}{\lambda} \right)

Generalized Cauchy

γ(h)=σ{1+(hλ)κ2}κ/κ2\gamma(h) = \sigma \left\{1 + \left(\frac{h}{\lambda} \right)^{\kappa_2} \right\}^{-\kappa / \kappa_2}

where σ\sigma, λ\lambda and κ\kappa are the sill, the range and shape parameters, Γ\Gamma is the gamma function, KκK_{\kappa} and JκJ_\kappa are both modified Bessel functions of order κ\kappa. In addition a nugget effect can be set that is there is a jump at the origin, i.e., γ(o)=ν+σ\gamma(o) = \nu + \sigma, where ν\nu is the nugget effect.

Value

This function returns the covariance function. Eventually, if dist is given, the covariance function is computed for each distance given by dist. If plot = TRUE, the covariance function is plotted.

Author(s)

Mathieu Ribatet

Examples

## 1- Calling covariance using fixed covariance parameters
covariance(nugget = 0, sill = 1, range = 1, smooth = 0.5, cov.mod = "whitmat")
covariance(nugget = 0, sill = 0.5, range = 1, smooth = 0.5, cov.mod = "whitmat",
  dist = seq(0,5, 0.2), plot = FALSE)

## 2- Calling covariance from a fitted model
##Define the coordinate of each location
n.site <- 30
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(30, locations, cov.mod = "whitmat", nugget = 0, range =
3, smooth = 1)

##Fit a max-stable model
fitted <- fitmaxstab(data, locations, "whitmat", nugget = 0)
covariance(fitted, ylim = c(0, 1))
covariance(nugget = 0, sill = 1, range = 3, smooth = 1, cov.mod = "whitmat", add =
TRUE, col = 3)
title("Whittle-Matern covariance function")
legend("topright", c("Theo.", "Fitted"), lty = 1, col = c(3,1), inset =
.05)

Estimates the penalty coefficient from the cross-validation criterion

Description

Estimates the penalty coefficient from the cross-validation criterion.

Usage

cv(y, x, knots, degree, plot = TRUE, n.points = 150, ...)

Arguments

y

The response vector.

x

A vector/matrix giving the values of the predictor variable(s). If x is a matrix, each row corresponds to one observation.

knots

A vector givint the coordinates of the knots.

degree

The degree of the penalized smoothing spline.

plot

Logical. If TRUE (default), the evolution of the CV score as the penalty increases is plotted.

n.points

A numeric giving the number of CV computations needed to produce the plot.

...

Options to be passed to the nlm function.

Details

For every linear smoother e.g. y^=Sλy\hat{y} = S_\lambda y, the cross-validation criterion consists in minimizing the following quantity:

CV(λ)=i=1n(yiy^i1Sλ,ii)2CV(\lambda) = \sum_{i=1}^n \left(\frac{y_i - \hat{y}_i}{1 - S_{\lambda,ii}} \right)^2

where λ\lambda is the penalty coefficient, nn the number of observations and Sλ,iiS_{\lambda,ii} the i-th diagonal element of the matrix SλS_\lambda.

Value

A list with components 'penalty', 'cv' and 'nlm.code' which give the location of the minimum, the value of the cross-validation criterion at that point and the code returned by the nlm function - useful to assess for convergence issues.

Author(s)

Mathieu Ribatet

References

Ruppert, D. Wand, M.P. and Carrol, R.J. (2003) Semiparametric Regression Cambridge Series in Statistical and Probabilistic Mathematics.

See Also

cv

Examples

n <- 200
x <- runif(n)
fun <- function(x) sin(3 * pi * x)
y <- fun(x) + rnorm(n, 0, sqrt(0.4))
knots <- quantile(x, prob = 1:(n/4) / (n/4 + 1))
cv(y, x, knots = knots, degree = 3)

Deviance Information Criterion

Description

This function computes the Deviance Information Criterion (DIC), and related quantities, which is a hierarchical modeling generalization of the Akaike Information Criterion. It is useful for Bayesian model selection.

Usage

DIC(object, ..., fun = "mean")

Arguments

object

An object of class latent — typically this will be the output of latent.

...

Optional arguments. Not implemented.

fun

A chararcter string giving the name of the function to be used to summarize the Markov chain. The default is to consider the posterior mean.

Details

The deviance is

D(θ)=2logπ(yθ),D(\theta) = -2 \log \pi(y \mid \theta),

where yy are the data, θ\theta are the unknown parameters of the models and π(yθ)\pi(y \mid \theta) is the likelihood function. Thus the expected deviance, a measure of how well the model fits the data, is given by

D=Eθ[D(θ)],\overline{D} = {\rm E}_{\theta}[D(\theta)],

while the effective number of parameters is

pD=DD(θ),p_D = \overline{D} - D(\theta^*),

where θ\theta^* is point estimate of the posterior distribution, e.g., the posterior mean. Finally the DIC is given by

DIC=pD+D.{\rm DIC} = p_D + \overline{D}.

In accordance with the AIC, models with smaller DIC should be preferred to models with larger DIC. Roughly speaking, differences of more than 10 might rule out the model with the higher DIC, differences between 5 and 10 are substantial.

Value

A vector of length 3 that returns the DIC, effective number of parameters eNoP and an estimate of the expected deviance Dbar.

Author(s)

Mathieu Ribatet

References

Spiegelhalter, D. J., Best, N. G., Carlin, B. P. and Van Der Linde, A. (2002) Bayesian measures of model complexity and fit. Journal of the Royal Statistical Society: Series B 64, 583–639.

See Also

AIC

Examples

## Generate realizations from the model
n.site <- 15
n.obs <- 35
coord <- cbind(lon = runif(n.site, -10, 10), lat = runif(n.site, -10 , 10))

gp.loc <- rgp(1, coord, "powexp", sill = 4, range = 20, smooth = 1)
gp.scale <- rgp(1, coord, "powexp", sill = 0.4, range = 5, smooth = 1)
gp.shape <- rgp(1, coord, "powexp", sill = 0.01, range = 10, smooth = 1)

locs <- 26 + 0.5 * coord[,"lon"] + gp.loc
scales <- 10 + 0.2 * coord[,"lat"] + gp.scale
shapes <- 0.15 + gp.shape

data <- matrix(NA, n.obs, n.site)
for (i in 1:n.site)
  data[,i] <- rgev(n.obs, locs[i], scales[i], shapes[i])

loc.form <- y ~ lon
scale.form <- y ~ lat
shape.form <- y ~ 1

hyper <- list()
hyper$sills <- list(loc = c(1,8), scale = c(1,1), shape = c(1,0.02))
hyper$ranges <- list(loc = c(2,20), scale = c(1,5), shape = c(1, 10))
hyper$smooths <- list(loc = c(1,1/3), scale = c(1,1/3), shape = c(1, 1/3))
hyper$betaMeans <- list(loc = rep(0, 2), scale = c(9, 0), shape = 0)
hyper$betaIcov <- list(loc = solve(diag(c(400, 100))),
                       scale = solve(diag(c(400, 100))),
                       shape = solve(diag(c(10), 1, 1)))

## We will use an exponential covariance function so the jump sizes for
## the shape parameter of the covariance function are null.
prop <- list(gev = c(2.5, 1.5, 0.3), ranges = c(40, 20, 20), smooths = c(0,0,0))
start <- list(sills = c(4, .36, 0.009), ranges = c(24, 17, 16), smooths
              = c(1, 1, 1),  beta = list(loc = c(26, 0), scale = c(10, 0),
                               shape = c(0.15)))

mc <- latent(data, coord, loc.form = loc.form, scale.form = scale.form,
             shape.form = shape.form, hyper = hyper, prop = prop, start = start,
             n = 500)
DIC(mc)

Computes distance between pairs of locations

Description

This function computes euclidean distance or distance vector for each pair of a set of spatial locations.

Usage

distance(coord, vec = FALSE)

Arguments

coord

A matrix representing the coordinates of each locations. Each row corresponds to one location.

vec

Logical. If FALSE (default), the euclidean distance is computed. Otherwise, “distance vectors” are returned.

Value

If vec = FALSE, this function returns a vector that gives the euclidean distance for each pair of locations. Otherwise, this is a matrix where each column correspond to one dimension - i.e. longitude, latitude, ...

Author(s)

Mathieu Ribatet

See Also

dist

Examples

coords <- cbind(seq(0, 10, by = 2), seq(10, 0, by = -2))
distance(coords)
distance(coords, vec = TRUE)

Plots the extremal coefficient

Description

Plots the extremal coefficient evolution as the coordinates evolves.

Usage

extcoeff(fitted, cov.mod, param, n = 200, xlab, ylab, ...)

Arguments

fitted

A object of class "maxstab". Most often, it will be the output of the function fitmaxstab. If missing, then cov.mod and param should be supplied.

cov.mod

A character string corresponding the the covariance model in the max-stable representation. Must be one of "gauss" for the Smith's model; or "whitmat", "cauchy" or "powexp" for the Whittle-Matern, the Cauchy and the Powered Exponential covariance family with the Schlather's model. May be missing if fitted is given.

param

Numeric vector of length 3. The parameters for the Smith's or Schlather model - i.e. c(cov11, cov12, cov22) or c(nugget, range, smooth). Please respect this order.

n

Numeric. n^2 corresponds to the total number of estimated extremal coefficients for the contour plot.

xlab, ylab

The x-axis and y-axis labels. May be missing.

...

Several options to be passed to the contour function.

Value

A plot.

Author(s)

Mathieu Ribatet

See Also

fitmaxstab

Examples

## 1- Random field generation
n.site <- 30
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

data <- rmaxstab(60, locations, cov.mod = "whitmat", nugget = 0, range =
3, smooth = 1)

## 2- Fit a max-stable processes
schlather <- fitmaxstab(data, locations, "whitmat", nugget = 0)

## 3- Plot the extremal coefficient
extcoeff(schlather)

Fit a copula-based model to spatial extremes

Description

This function fits various copula-based models to spatial extremes data sets.

Usage

fitcopula(data, coord, copula = "gaussian", cov.mod = "whitmat",
loc.form, scale.form, shape.form, marg.cov = NULL, temp.cov = NULL,
temp.form.loc = NULL, temp.form.scale = NULL, temp.form.shape = NULL,
..., start, control = list(maxit = 10000), method = "Nelder", std.err =
TRUE, warn = TRUE, corr = FALSE)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

copula

A character string. Must be one of "gaussian" and "student" for a Gaussian and Student copula.

cov.mod

A character string corresponding to the correlation function family used in the copula. Must be one of "whitmat", "cauchy", "powexp", "bessel" or "caugen" for the Whittle-Matern, the Cauchy, the Powered Exponential, the Bessel and the Generalized Cauchy correlation families.

loc.form, scale.form, shape.form

R formulas defining the spatial linear model for the GEV parameters. May be missing. See section Details of function fitmaxstab.

marg.cov

Matrix with named columns giving additional covariates for the GEV parameters. If NULL, no extra covariates are used.

temp.cov

Matrix with names columns giving additional *temporal* covariates for the GEV parameters. If NULL, no temporal trend are assume for the GEV parameters — see section Details of function fitmaxstab.

temp.form.loc, temp.form.scale, temp.form.shape

R formulas defining the temporal trends for the GEV parameters. May be missing. See section Details of function fitmaxstab.

...

Several arguments to be passed to the optim, nlm or nlminb functions.

start

A named list giving the initial values for the parameters over which the pairwise likelihood is to be minimized. If start is omitted the routine attempts to find good starting values - but might fail.

control

A list giving the control parameters to be passed to the optim function.

method

The method used for the numerical optimisation procedure. Must be one of BFGS, Nelder-Mead, CG, L-BFGS-B, SANN, nlm or nlminb. See optim for details. Please note that passing nlm or nlminb will use the nlm or nlminb functions instead of optim.

std.err

Logical. Should the standard errors be computed ? The default is to return the standard errors, i.e., std.err = TRUE.

warn

Logical. If TRUE (default), users are warned if the log-likelihood is infinite at starting values and/or problems arised while computing the standard errors.

corr

Logical. If TRUE (non default), the asymptotic correlation matrix is computed.

Value

This function returns a object of class copula.

Warning

This function does not use max-stable copula and the use of non max-stable copula for modelling spatial extreme is highly questionable. This function was mainly implemented for educational purposes and not for concrete modelling purposes.

Author(s)

Mathieu Ribatet

References

Davison, A.C., Padoan, S.A., Ribatet, M. (2010) Statistical Modelling of Spatial Extremes. Submitted to Statistical Science.

See Also

fitmaxstab, latent

Examples

## Not run: 
n.site <- 30
n.obs <- 50

coord <- matrix(runif(2 * n.site, -10, 10), ncol = 2)
colnames(coord) <- c("lon", "lat")

## Generate data from a Gaussian copula model
data <- rcopula(n.obs, coord, "gaussian", "powexp", nugget = 0, range = 4, smooth = 1.2)

## Transform the margins to GEV
locs <- -5 + coord[,"lon"] / 10
scales <- 10 + coord[,"lat"] / 2
shapes <- rep(0.2, n.site)

for (i in 1:n.site)
  data[,i] <- frech2gev(data[,i], locs[i], scales[i], shapes[i])

## Fit a Gaussian copula model

## 1. Define trend surfaces
loc.form <- y ~ lon
scale.form <- y ~ lat
shape.form <- y ~ 1

## 2. Fit
M0 <- fitcopula(data, coord, "gaussian", "powexp", loc.form, scale.form,
                shape.form, nugget = 0)

## End(Not run)

Estimates the covariance function for the Schlather's model

Description

Estimates the covariance function for the Schlather's model using non-parametric estimates of the pairwise extremal coefficients.

Usage

fitcovariance(data, coord, cov.mod, marge = "emp", control = list(),
..., start, weighted = TRUE)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

cov.mod

A character string corresponding the the covariance model in the Schlather's model. Must be one of "whitmat", "cauchy", "powexp", "bessel" or "caugen" for the Whittle-Matern, the Cauchy, the Powered Exponential, the Bessel and the Generalized Cauchy correlation families.

marge

Character string specifying how margins are transformed to unit Frechet. Must be one of "emp", "frech" or "mle" - see function fitextcoeff.

control

The control arguments to be passed to the optim function.

...

Optional arguments to be passed to the optim function.

start

A named list giving the initial values for the parameters over which the weighted sum of square is to be minimized. If start is omitted the routine attempts to find good starting values.

weighted

Logical. Should weighted least squares be used?

Details

The fitting procedure is based on weighted least squares. More precisely, the fitting criteria is to minimize:

i,j(θ~i,jθ^i,jsi,j)2\sum_{i,j} \left(\frac{\tilde{\theta}_{i,j} - \hat{\theta}_{i,j}}{s_{i,j}}\right)^2

where θ~i,j\tilde{\theta}_{i,j} is a non parametric estimate of the extremal coefficient related to location i and j, θ^i,j\hat{\theta}_{i,j} is the fitted extremal coefficient derived from the Schlather's model and si,js_{i,j} are the standard errors related to the estimates θ~i,j\tilde{\theta}_{i,j}.

Value

An object of class maxstab.

Author(s)

Mathieu Ribatet

References

Smith, R. L. (1990) Max-stable processes and spatial extremes. Unpublished manuscript.

See Also

fitcovmat, fitmaxstab, fitextcoeff

Examples

n.site <- 50
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulating a max-stable process using RandomFields
##This is the Schlather's approach
data <- rmaxstab(50, locations, cov.mod = "whitmat", nugget = 0, range =
30, smooth = 1)

fitcovariance(data, locations, "whitmat")

Estimates the covariance matrix for the Smith's model

Description

Estimates the covariance matrix for the Smith's model using non-parametric estimates of the pairwise extremal coefficients.

Usage

fitcovmat(data, coord, marge = "emp", iso = FALSE, control = list(),
..., start, weighted = TRUE)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

marge

Character string specifying how margins are transformed to unit Frechet. Must be one of "emp", "frech" or "mle" - see function fitextcoeff.

iso

Logical. If TRUE, isotropy is supposed. Otherwise (default), anisotropy is allowed.

control

The control arguments to be passed to the optim function.

...

Optional arguments to be passed to the optim function.

start

A named list giving the initial values for the parameters over which the weighted sum of square is to be minimized. If start is omitted the routine attempts to find good starting values.

weighted

Logical. Should weighted least squares be used?

Details

The fitting procedure is based on weighted least squares. More precisely, the fitting criteria is to minimize:

i,j(θ~i,jθ^i,jsi,j)2\sum_{i,j} \left(\frac{\tilde{\theta}_{i,j} - \hat{\theta}_{i,j}}{s_{i,j}}\right)^2

where θ~i,j\tilde{\theta}_{i,j} is a non parametric estimate of the extremal coefficient related to location i and j, θ^i,j\hat{\theta}_{i,j} is the fitted extremal coefficient derived from the Smith's model and si,js_{i,j} are the standard errors related to the estimates θ~i,j\tilde{\theta}_{i,j}.

Value

An object of class maxstab.

Author(s)

Mathieu Ribatet

References

Smith, R. L. (1990) Max-stable processes and spatial extremes. Unpublished manuscript.

See Also

fitcovariance, fitmaxstab, fitextcoeff

Examples

n.site <- 50
n.obs <- 100
locations <- matrix(runif(2*n.site, 0, 40), ncol = 2)
colnames(locations) <- c("lon", "lat")

## Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(50, locations, cov.mod = "gauss", cov11 = 200, cov12 =
0, cov22 = 200)

fitcovmat(data, locations)

##Force an isotropic model
fitcovmat(data, locations, iso = TRUE)

Non parametric estimators of the extremal coefficient function

Description

Estimates non parametrically the extremal coefficient function.

Usage

fitextcoeff(data, coord, ..., estim = "ST", marge = "emp", prob = 0,
plot = TRUE, loess = TRUE, method = "BFGS", std.err = TRUE, xlab,
ylab, angles = NULL, identify = FALSE)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

...

Additional options to be passed to the plot function.

estim

Character string specifying the estimator to be used. Must be one of "ST" (Schlather and Tawn) or "Smith".

marge

Character string specifying how margins are transformed to unit Frechet. Must be one of "emp", "mle" or "none" - see Details

prob

The probability related to the threshold. Only useful with the ST estimator.

plot

Logical. If TRUE (default), the extremal coefficient function is plotted

loess

If TRUE (default), a local polynomial curve is plotted - see function loess.

method

The optimizer used when fitting the GEV distribution to data. See function gevmle.

std.err

Logical. If TRUE, standard errors are computed. Note that standard errors are not available with the "ST" estimator.

xlab, ylab

The x-axis and y-axis labels. May be missing.

angles

A numeric vector. A partition of the interval (π,π)(-\pi, \pi) to help detecting anisotropy.

identify

Logical. If TRUE, users can use the identify function to identify pairs of stations on the plot.

Details

During the estimation procedure, data need to be transformed to unit Frechet margins firts. This can be done in two different ways ; by using the empirical CDF or the GEV ML estimates.

If marge = "emp", then the data are transformed using the following relation:

zi=1log(F(yi))z_i = - \frac{1}{\log (F(y_i))}

where yiy_i are the observations available at location ii, FF is the empirical CDF and ziz_i are the observations transformed to unit Frechet scale.

If marge = "mle", then the data are transformed using the MLE of the GEV distribution - see function gev2frech.

Lastly, if data are already supposed to be unit Frechet, then no transformation is performed if one passed the option marge = "frech".

If data are already componentwise maxima, prob should be zero. Otherwise, users have to define a threshold zz (large enough) where univariate extreme value arguments are relevant. We define prob such that Pr[Zz]=prob\Pr[Z \leq z] = prob.

Value

Plots the extremal coefficient function and returns the points used for the plot. If loess = TRUE, the output is a list with argument "ext.coeff" and "loess".

Author(s)

Mathieu Ribatet

References

Schlather, M. and Tawn, J. A. (2003) A dependence measure for multivariate and spatial extreme values: Properties and inference. Biometrika 90(1):139–156.

Smith, R. L. (1990) Max-stable processes and spatial extremes. Unpublished manuscript.

See Also

madogram

Examples

n.site <- 30
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(50, locations, cov.mod = "gauss", cov11 = 10, cov12 =
40, cov22 = 220)

##Plot the extremal coefficient function
op <- par(mfrow=c(1,2))
fitextcoeff(data, locations, estim = "Smith")
fitextcoeff(data, locations, angles = seq(-pi, pi, length = 4), estim = "Smith")
par(op)

Fits a max-stable process to data

Description

This function fits max-stable processes to data using pairwise likelihood. Two max-stable characterisations are available: the Smith and Schlather representations.

Usage

fitmaxstab(data,  coord, cov.mod, loc.form, scale.form,
shape.form, marg.cov = NULL, temp.cov = NULL, temp.form.loc = NULL,
temp.form.scale = NULL, temp.form.shape = NULL, iso = FALSE, ...,
fit.marge = FALSE, warn = TRUE, method = "Nelder", start, control =
list(), weights = NULL, corr = FALSE, check.grad
= FALSE)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

cov.mod

A character string corresponding to the covariance model in the max-stable representation. Must be one of "gauss" for the Smith's model; "whitmat", "cauchy", "powexp", "bessel" or "caugen" for the Whittle-Matern, the Cauchy, the Powered Exponential, the Bessel and the Generalized Cauchy correlation families with the Schlather's model; "brown" for Brown-Resnick processes. The geometric Gaussian and Extremal-t models with a Whittle-Matern correlation function can be fitted by passing respectively "gwhitmat" or "twhitmat". Other correlation function families are considered in a similar way.

loc.form, scale.form, shape.form

R formulas defining the spatial linear model for the GEV parameters. May be missing. See section Details.

marg.cov

Matrix with named columns giving additional covariates for the GEV parameters. If NULL, no extra covariates are used.

temp.cov

Matrix with names columns giving additional *temporal* covariates for the GEV parameters. If NULL, no temporal trend are assume for the GEV parameters — see section Details.

temp.form.loc, temp.form.scale, temp.form.shape

R formulas defining the temporal trends for the GEV parameters. May be missing. See section Details.

iso

Logical. If TRUE an isotropic model is fitted to data. Otherwise (default), anisotropy is allowed. Currently, this is only implemented for the Smith's model.

...

Several arguments to be passed to the optim, nlm or nlminb functions. See section details.

fit.marge

Logical. If TRUE, the GEV parameters are estimated pointwise or using the formulas given by loc.form, scale.form and shape.form. If FALSE, observations are supposed to be unit Frechet distributed. Note that when formulas are given, fit.marge is automatically set to TRUE.

warn

Logical. If TRUE (default), users are warned if the log-likelihood is infinite at starting values and/or problems arised while computing the standard errors.

method

The method used for the numerical optimisation procedure. Must be one of BFGS, Nelder-Mead, CG, L-BFGS-B, SANN, nlm or nlminb. See optim for details. Please note that passing nlm or nlminb will use the nlm or nlminb functions instead of optim.

start

A named list giving the initial values for the parameters over which the pairwise likelihood is to be minimized. If start is omitted the routine attempts to find good starting values - but might fail.

control

A list giving the control parameters to be passed to the optim function.

weights

A numeric vector specifying the weights in the pairwise likelihood - and so has length the number of pairs. If NULL (default), no weighting scheme is used.

corr

Logical. If TRUE (non default), the asymptotic correlation matrix is computed.

check.grad

Logical. If TRUE (non default), the analytic gradient of the pairwise likelihood will be compared to the numerical one. Such a checking might be usefull for ill-conditionned situation diagnosis.

Details

As spatial data often deal with a large number of locations, it is impossible to write analytically the joint distribution. Consequently, the fitting procedure substitutes the "full likelihood" for the pairwise likelihood.

Let define Li,j(xi,j,θ)L_{i,j}(x_{i,j}, \theta) the likelihood for site ii and jj, where i=1,,N1i = 1, \dots, N-1, j=i+1,,Nj = i+1, \dots, N, NN is the number of site within the region and xi,jx_{i,j} are the joint observations for site ii and jj. Then the pairwise likelihood PL(θ)PL(\theta) is defined by:

P=logPL(θ)=i=1N1j=i+1NlogLi,j(xi,j,θ)\ell_P = \log PL(\theta) = \sum_{i = 1}^{N-1} \sum_{j=i+1}^{N} \log L_{i,j} (x_{i,j}, \theta)

As pairwise likelihood is an approximation of the “full likelihood”, standard errors cannot be computed directly by the inverse of the Fisher information matrix. Instead, a sandwich estimate must be used to account for model mispecification e.g.

θ^N(θ,H1JH1)\hat{\theta} \sim N(\theta, H^{-1} J H^{-1})

where HH is the Fisher information matrix (computed from the misspecified model) and JJ is the variance of the score function.

There are two different kind of covariates : "spatial" and "temporal".

A "spatial" covariate may have different values accross station but does not depend on time. For example the coordinates of the stations are obviously "spatial". These "spatial" covariates should be used with the marg.cov and loc.form, scale.form, shape.form.

A "temporal" covariates may have different values accross time but does not depend on space. For example the years where the annual maxima were recorded is "temporal". These "temporal" covariates should be used with the temp.cov and temp.form.loc, temp.form.scale, temp.form.shape.

As a consequence note that marg.cov must have K rows (K being the number of sites) while temp.cov must have n rows (n being the number of observations).

Value

This function returns a object of class maxstab. Such objects are list with components:

fitted.values

A vector containing the estimated parameters.

std.err

A vector containing the standard errors.

fixed

A vector containing the parameters of the model that have been held fixed.

param

A vector containing all parameters (optimised and fixed).

deviance

The (pairwise) deviance at the maximum pairwise likelihood estimates.

corr

The correlation matrix.

convergence, counts, message

Components taken from the list returned by optim - for the mle method.

data

The data analysed.

model

The max-stable characterisation used.

fit.marge

A logical that specifies if the GEV margins were estimated.

cov.fun

The estimated covariance function - for the Schlather model only.

extCoeff

The estimated extremal coefficient function.

cov.mod

The covariance model for the spatial structure.

Warning

When using reponse surfaces to model spatially the GEV parameters, the likelihood is pretty rough so that the general purpose optimization routines may fail. It is your responsability to check if the numerical optimization succeeded or not. I tried, as best as I can, to provide warning messages if the optimizers failed but in some cases, no warning will appear!

Author(s)

Mathieu Ribatet

References

Cox, D. R. and Reid, N. (2004) A note on pseudo-likelihood constructed from marginal densities. Biometrika 91, 729–737.

Demarta, S. and McNeil, A. (2005) The t copula and Related Copulas International Statistical Review 73, 111-129.

Gholam–Rezaee, M. (2009) Spatial extreme value: A composite likelihood. PhD Thesis. Ecole Polytechnique Federale de Lausanne.

Kabluchko, Z., Schlather, M. and de Haan, L. (2009) Stationary max-stable fields associated to negative definite functions Annals of Probability 37:5, 2042–2065.

Padoan, S. A. (2008) Computational Methods for Complex Problems in Extreme Value Theory. PhD Thesis. University of Padova.

Padoan, S. A., Ribatet, M. and Sisson, S. A. (2010) Likelihood-based inference for max-stable processes. Journal of the American Statistical Association (Theory and Methods) 105:489, 263-277.

Schlather, M. (2002) Models for Stationary Max-Stable Random Fields. Extremes 5:1, 33–44.

Smith, R. L. (1990) Max-stable processes and spatial extremes. Unpublished.

Examples

## Not run: 
##Define the coordinate of each location
n.site <- 30
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(40, locations, cov.mod = "whitmat", nugget = 0, range = 3,
smooth = 0.5)

##Now define the spatial model for the GEV parameters
param.loc <- -10 + 2 * locations[,2]
param.scale <- 5 + 2 * locations[,1] + locations[,2]^2
param.shape <- rep(0.2, n.site)

##Transform the unit Frechet margins to GEV
for (i in 1:n.site)
  data[,i] <- frech2gev(data[,i], param.loc[i], param.scale[i],
param.shape[i])

##Define a model for the GEV margins to be fitted
##shape ~ 1 stands for the GEV shape parameter is constant
##over the region
loc.form <- loc ~ lat
scale.form <- scale ~ lon + I(lat^2)
shape.form <- shape ~ 1

##Fit a max-stable process using the Schlather's model
fitmaxstab(data, locations, "whitmat", loc.form, scale.form,
           shape.form)

## Model without any spatial structure for the GEV parameters
## Be careful this could be *REALLY* time consuming
fitmaxstab(data, locations, "whitmat")

##  Fixing the smooth parameter of the Whittle-Matern family
##  to 0.5 - e.g. considering exponential family. We suppose the data
##  are unit Frechet here.
fitmaxstab(data, locations, "whitmat", smooth = 0.5, fit.marge = FALSE)

##  Fitting a penalized smoothing splines for the margins with the
##     Smith's model
data <- rmaxstab(40, locations, cov.mod = "gauss", cov11 = 100, cov12 =
                 25, cov22 = 220)

##     And transform it to ordinary GEV margins with a non-linear
##     function
fun <- function(x)
  2 * sin(pi * x / 4) + 10
fun2 <- function(x)
  (fun(x) - 7 ) / 15

param.loc <- fun(locations[,2])
param.scale <- fun(locations[,2])
param.shape <- fun2(locations[,1])

##Transformation from unit Frechet to common GEV margins
for (i in 1:n.site)
  data[,i] <- frech2gev(data[,i], param.loc[i], param.scale[i],
param.shape[i])

##Defining the knots, penalty, degree for the splines
n.knots <- 5
knots <- quantile(locations[,2], prob = 1:n.knots/(n.knots+1))
knots2 <- quantile(locations[,1], prob = 1:n.knots/(n.knots+1))

##Be careful the choice of the penalty (i.e. the smoothing parameter)
##may strongly affect the result Here we use p-splines for each GEV
##parameter - so it's really CPU demanding but one can use 1 p-spline
##and 2 linear models.
##A simple linear model will be clearly faster...
loc.form <- y ~ rb(lat, knots = knots, degree = 3, penalty = .5)
scale.form <- y ~ rb(lat, knots = knots, degree = 3, penalty = .5)
shape.form <- y ~ rb(lon, knots = knots2, degree = 3, penalty = .5)

fitted <- fitmaxstab(data, locations, "gauss", loc.form, scale.form, shape.form,
                     control = list(ndeps = rep(1e-6, 24), trace = 10),
                     method = "BFGS")
fitted
op <- par(mfrow=c(1,3))
plot(locations[,2], param.loc, col = 2, ylim = c(7, 14),
     ylab = "location parameter", xlab = "latitude")
plot(fun, from = 0, to = 10, add = TRUE, col = 2)
points(locations[,2], predict(fitted)[,"loc"], col = "blue", pch = 5)
new.data <- cbind(lon = seq(0, 10, length = 100), lat = seq(0, 10, length = 100))
lines(new.data[,1], predict(fitted, new.data)[,"loc"], col = "blue")
legend("topleft", c("true values", "predict. values", "true curve", "predict. curve"),
       col = c("red", "blue", "red", "blue"), pch = c(1, 5, NA, NA), inset = 0.05,
       lty = c(0, 0, 1, 1), ncol = 2)

plot(locations[,2], param.scale, col = 2, ylim = c(7, 14),
     ylab = "scale parameter", xlab = "latitude")
plot(fun, from = 0, to = 10, add = TRUE, col = 2)
points(locations[,2], predict(fitted)[,"scale"], col = "blue", pch = 5)
lines(new.data[,1], predict(fitted, new.data)[,"scale"], col = "blue")
legend("topleft", c("true values", "predict. values", "true curve", "predict. curve"),
       col = c("red", "blue", "red", "blue"), pch = c(1, 5, NA, NA), inset = 0.05,
       lty = c(0, 0, 1, 1), ncol = 2)

plot(locations[,1], param.shape, col = 2,
     ylab = "shape parameter", xlab = "longitude")
plot(fun2, from = 0, to = 10, add = TRUE, col = 2)
points(locations[,1], predict(fitted)[,"shape"], col = "blue", pch = 5)
lines(new.data[,1], predict(fitted, new.data)[,"shape"], col = "blue")
legend("topleft", c("true values", "predict. values", "true curve", "predict. curve"),
       col = c("red", "blue", "red", "blue"), pch = c(1, 5, NA, NA), inset = 0.05,
       lty = c(0, 0, 1, 1), ncol = 2)
par(op)

## End(Not run)

MLE for a spatial GEV model

Description

This function derives the MLE of a spatial GEV model.

Usage

fitspatgev(data, covariables, loc.form, scale.form, shape.form,
temp.cov = NULL, temp.form.loc = NULL, temp.form.scale = NULL,
temp.form.shape = NULL, ..., start, control = list(maxit = 10000),
method = "Nelder", warn = TRUE, corr = FALSE)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

covariables

Matrix with named columns giving required covariates for the GEV parameter models.

loc.form, scale.form, shape.form

R formulas defining the spatial models for the GEV parameters. See section Details.

temp.cov

Matrix with names columns giving additional *temporal* covariates for the GEV parameters. If NULL, no temporal trend are assume for the GEV parameters — see section Details.

temp.form.loc, temp.form.scale, temp.form.shape

R formulas defining the temporal trends for the GEV parameters. May be missing. See section Details.

start

A named list giving the initial values for the parameters over which the pairwise likelihood is to be minimized. If start is omitted the routine attempts to find good starting values - but might fail.

...

Several arguments to be passed to the optim functions. See section details.

control

The control argument to be passed to the optim function.

method

The method used for the numerical optimisation procedure. Must be one of BFGS, Nelder-Mead, CG, L-BFGS-B or SANN. See optim for details.

warn

Logical. If TRUE (default), users will be warned if the starting values lie in a zero density region.

corr

Logical. If TRUE, the asymptotic correlation matrix is computed.

Details

A kind of "spatial" GEV model can be defined by using response surfaces for the GEV parameters. For instance, the GEV location parameters are defined through the following equation:

μ=Xμβμ\mu = X_\mu \beta_\mu

where XμX_\mu is the design matrix and βμ\beta_\mu is the vector parameter to be estimated. The GEV scale and shape parameters are defined accordingly to the above equation.

The log-likelihood for the GEV spatial model is consequently defined as follows:

(β)=i=1n.sitej=1n.obslogf(yi,j;θi)\ell(\beta) = \sum_{i=1}^{n.site} \sum_{j=1}^{n.obs} \log f(y_{i,j}; \theta_i)

where θi\theta_i is the vector of the GEV parameters for the ii-th site.

Most often, there will be some dependence between stations. However, it can be seen from the log-likelihood definition that we supposed that the stations are mutually independent. Consequently, to get reliable standard error estimates, these standard errors are estimated with their sandwich estimates.

There are two different kind of covariates : "spatial" and "temporal".

A "spatial" covariate may have different values accross station but does not depend on time. For example the coordinates of the stations are obviously "spatial". These "spatial" covariates should be used with the marg.cov and loc.form, scale.form, shape.form.

A "temporal" covariates may have different values accross time but does not depend on space. For example the years where the annual maxima were recorded is "temporal". These "temporal" covariates should be used with the temp.cov and temp.form.loc, temp.form.scale, temp.form.shape.

As a consequence note that marg.cov must have K rows (K being the number of sites) while temp.cov must have n rows (n being the number of observations).

Value

An object of class spatgev. Namely, this is a list with the following arguments:

fitted.values

The parameter estimates.

param

All the parameters e.g. parameter estimates and fixed parameters.

std.err

The standard errors.

var.cov

The asymptotic MLE variance covariance matrix.

counts, message, convergence

Some information about the optimization procedure.

logLik, deviance

The log-likelihood and deviance values.

loc.form, scale.form, shape.form

The formulas defining the spatial models for the GEV parameters.

covariables

The covariables used for the spatial models.

ihessian

The inverse of the Hessian matrix of the negative log-likelihood.

jacobian

The variance covariance matrix of the score.

Author(s)

Mathieu Ribatet

Examples

## 1- Simulate a max-stable random field
n.site <- 35
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")
data <- rmaxstab(50, locations, cov.mod = "whitmat", nugget = 0, range =
  3, smooth = 0.5)

## 2- Transformation to non unit Frechet margins
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)
  data[,i] <- frech2gev(data[,i], param.loc[i], param.scale[i],
param.shape[i])

## 3- Fit a ''spatial GEV'' mdoel to data with the following models for
##    the GEV parameters
form.loc <- loc ~ lat
form.scale <- scale ~ lon
form.shape <- shape ~ 1

fitspatgev(data, locations, form.loc, form.scale, form.shape)

Computes the F-madogram

Description

Computes the F-madogram for max-stable processes.

Usage

fmadogram(data, coord, fitted, n.bins, which = c("mado", "ext"), xlab,
ylab, col = c(1, 2), angles = NULL, marge = "emp", add = FALSE, xlim =
c(0, max(dist)), ...)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

fitted

An object of class maxstab - usually the output of the fitmaxstab function. May be missing.

n.bins

The number of bins to be used. If missing, pairwise F-madogram estimates will be computed.

which

A character vector of maximum size 2. It specifies if the madogram and/or the extremal coefficient functions have to be plotted.

xlab, ylab

The x-axis and y-axis labels. May be missing. Note that ylab must have the same length has which.

col

The colors used for the points and optionnaly the fitted curve.

angles

A numeric vector. A partition of the interval (π,π)(-\pi, \pi) to help detecting anisotropy.

marge

Character string. If 'emp', the probabilities of non exceedances are estimated using the empirical CDF. If 'mle' (default), maximum likelihood estimates are used.

add

Logical. If TRUE, the plot is added to the current figure; otherwhise (default) a new plot is computed.

xlim

A numeric vector of length 2 specifying the x coordinate range.

...

Additional options to be passed to the plot function.

Details

Let Z(x)Z(x) be a stationary process. The F-madogram is defined as follows:

ν(h)=12E[F(Z(x+h))F(Z(x))]\nu(h) = \frac{1}{2}\mbox{E}\left[|F(Z(x+h)) - F(Z(x))| \right]

The extremal coefficient θ(h)\theta(h) satisfies:

θ(h)=1+2ν(h)12ν(h)\theta(h) = \frac{1 + 2 \nu(h)}{1 - 2 \nu(h)}

Value

A graphic and (invisibly) a matrix with the lag distances, the F-madogram and extremal coefficient estimates.

Author(s)

Mathieu Ribatet

References

Cooley, D., Naveau, P. and Poncet, P. (2006) Variograms for spatial max-stable random fields. Dependence in Probability and Statistics, 373–390.

See Also

madogram, lmadogram

Examples

n.site <- 15
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(40, locations, cov.mod = "whitmat", nugget = 0, range = 1,
smooth = 2)

##Compute the F-madogram
fmadogram(data, locations)

##Compare the F-madogram with a fitted max-stable process
fitted <- fitmaxstab(data, locations, "whitmat", nugget = 0)
fmadogram(fitted = fitted, which = "ext")

Estimates the penalty coefficient from the generalized cross-validation criterion

Description

Estimates the penalty coefficient from the generalized cross-validation criterion.

Usage

gcv(y, x, knots, degree, plot = TRUE, n.points = 150, ...)

Arguments

y

The response vector.

x

A vector/matrix giving the values of the predictor variable(s). If x is a matrix, each row corresponds to one observation.

knots

A vector givint the coordinates of the knots.

degree

The degree of the penalized smoothing spline.

plot

Logical. If TRUE (default), the evolution of the CV score as the penalty increases is plotted.

n.points

A numeric giving the number of CV computations needed to produce the plot.

...

Options to be passed to the nlm function.

Details

For every linear smoother e.g. y^=Sλy\hat{y} = S_\lambda y, the cross-validation criterion consists in minimizing the following quantity:

GCV(λ)=nyy^2(ntr(Sλ))2GCV(\lambda) = \frac{n ||y - \hat{y}||^2}{(n - tr(S_\lambda))^2}

where λ\lambda is the penalty coefficient, nn the number of observations and tr(Sλ)tr(S_\lambda) is the trace of the matrix SλS_\lambda.

Value

A list with components 'penalty', 'gcv' and 'nlm.code' which give the location of the minimum, the value of the cross-validation criterion at that point and the code returned by the link{nlm} function - useful to assess for convergence issues.

Author(s)

Mathieu Ribatet

References

Ruppert, D. Wand, M.P. and Carrol, R.J. (2003) Semiparametric Regression Cambridge Series in Statistical and Probabilistic Mathematics.

See Also

cv

Examples

n <- 200
x <- runif(n)
fun <- function(x) sin(3 * pi * x)
y <- fun(x) + rnorm(n, 0, sqrt(0.4))
knots <- quantile(x, prob = 1:(n/4) / (n/4 + 1))
gcv(y, x, knots = knots, degree = 3)

The Generalized Extreme Value Distribution

Description

Density, distribution function, quantile function and random generation for the GP distribution with location equal to 'loc', scale equal to 'scale' and shape equal to 'shape'.

Usage

rgev(n, loc = 0, scale = 1, shape = 0)
pgev(q, loc = 0, scale = 1, shape = 0, lower.tail = TRUE)
qgev(p, loc = 0, scale = 1, shape = 0, lower.tail = TRUE)
dgev(x, loc = 0, scale = 1, shape = 0, log = FALSE)

Arguments

x, q

vector of quantiles.

p

vector of probabilities.

n

number of observations.

loc

vector of the location parameters.

scale

vector of the scale parameters.

shape

a numeric of the shape parameter.

lower.tail

logical; if TRUE (default), probabilities are Pr[Xx]\Pr[ X \le x], otherwise, Pr[X>x]\Pr[X > x].

log

logical; if TRUE, probabilities p are given as log(p).

Value

If 'loc', 'scale' and 'shape' are not specified they assume the default values of '0', '1' and '0', respectively.

The GEV distribution function for loc = uu, scale = σ\sigma and shape = ξ\xi is

G(x)=exp[{1+ξxuσ}1/ξ]G(x) = \exp\left[-\left\{1 + \xi \frac{x - u}{\sigma} \right\}^{-1 / \xi} \right]

for 1+ξ(xu)/σ>01 + \xi ( x - u ) / \sigma > 0 and x>ux > u, where σ>0\sigma > 0. If ξ=0\xi = 0, the distribution is defined by continuity corresponding to the Gumbel distribution.

Examples

dgev(0.1)
rgev(100, 1, 2, 0.2)
qgev(seq(0.1, 0.9, 0.1), 1, 0.5, -0.2)
pgev(12.6, 2, 0.5, 0.1)

The Generalized Pareto Distribution

Description

Density, distribution function, quantile function and random generation for the GP distribution with location equal to 'loc', scale equal to 'scale' and shape equal to 'shape'.

Usage

rgpd(n, loc = 0, scale = 1, shape = 0)
pgpd(q, loc = 0, scale = 1, shape = 0, lower.tail = TRUE, lambda = 0)
qgpd(p, loc = 0, scale = 1, shape = 0, lower.tail = TRUE, lambda = 0)
dgpd(x, loc = 0, scale = 1, shape = 0, log = FALSE)

Arguments

x, q

vector of quantiles.

p

vector of probabilities.

n

number of observations.

loc

vector of the location parameters.

scale

vector of the scale parameters.

shape

a numeric of the shape parameter.

lower.tail

logical; if TRUE (default), probabilities are Pr[Xx]\Pr[ X \le x], otherwise, Pr[X>x]\Pr[X > x].

log

logical; if TRUE, probabilities p are given as log(p).

lambda

a single probability - see the "value" section.

Value

If 'loc', 'scale' and 'shape' are not specified they assume the default values of '0', '1' and '0', respectively.

The GP distribution function for loc = uu, scale = σ\sigma and shape = ξ\xi is

G(x)=1[1+ξ(xu)σ]1/ξG(x) = 1 - \left[ 1 + \frac{\xi (x - u )}{ \sigma } \right] ^ { - 1 / \xi}

for 1+ξ(xu)/σ>01 + \xi ( x - u ) / \sigma > 0 and x>ux > u, where σ>0\sigma > 0. If ξ=0\xi = 0, the distribution is defined by continuity corresponding to the exponential distribution.

By definition, the GP distribution models exceedances above a threshold. In particular, the GG function is a suited candidate to model

Pr[XxX>u]=1G(x)\Pr\left[ X \geq x | X > u \right] = 1 - G(x)

for uu large enough.

However, it may be usefull to model the "non conditional" quantiles, that is the ones related to Pr[Xx]\Pr[ X \leq x]. Using the conditional probability definition, one have :

Pr[Xx]=(1λ)(1+ξxuσ)1/ξ\Pr\left[ X \geq x \right] = \left(1 - \lambda\right) \left( 1 + \xi \frac{x - u}{\sigma}\right)^{-1/\xi}

where λ=Pr[Xu]\lambda = \Pr[ X \leq u].

When λ=0\lambda = 0, the "conditional" distribution is equivalent to the "non conditional" distribution.

Examples

dgpd(0.1)
rgpd(100, 1, 2, 0.2)
qgpd(seq(0.1, 0.9, 0.1), 1, 0.5, -0.2)
pgpd(12.6, 2, 0.5, 0.1)
##for non conditional quantiles
qgpd(seq(0.9, 0.99, 0.01), 1, 0.5, -0.2, lambda = 0.9)
pgpd(2.6, 2, 2.5, 0.25, lambda = 0.5)

Transforms GEV data to unit Frechet ones and vice versa

Description

Transforms GEV data to unit Frechet ones and vice versa

Usage

gev2frech(x, loc, scale, shape, emp = FALSE)
frech2gev(x, loc, scale, shape)

Arguments

x

The data to be transformed to unit Frechet or ordinary GEV margins

loc, scale, shape

The location, scale and shape parameters of the GEV.

emp

Logical. If TRUE, data are transformed to unit Frechet margins using the empirical CDF.

Details

If Y is a random variable with a GEV distribution with location μ\mu, scale σ\sigma and shape ξ\xi. Then,

Z=[1+ξYμσ]+1/ξZ = \left[1 + \xi \frac{Y - \mu}{\sigma} \right]_+^{1/\xi}

is unit Frechet distributed - where x+=max(0,x)x_+ = \max(0, x).

If Z is a unit Frechet random variable. Then,

Y=μ+σZ+ξ1ξY = \mu + \sigma \frac{Z_+^{\xi} - 1}{\xi}

is unit GEV distributed with location, scale and shape parameters equal to μ\mu, σ\sigma and ξ\xi respectively.

Value

Returns a numeric vector with the same length of x

Author(s)

Mathieu Ribatet

Examples

x <- c(2.2975896, 1.6448808, 1.3323833, -0.4464904, 2.2737603,
    -0.2581876, 9.5184398, -0.5899699, 0.4974283, -0.8152157)
y <- gev2frech(x, 1, 2, .2)
y
frech2gev(y, 1, 2, .2)

Simple kriging interpolation

Description

This function interpolates a zero mean Gaussian random field using the simple kriging predictor.

Usage

kriging(data, data.coord, krig.coord, cov.mod = "whitmat", sill, range,
smooth, smooth2 = NULL, grid = FALSE, only.weights = FALSE)

Arguments

data

A numeric vector or matrix. If data is a matrix then the simple kriging predictor is given for each realisation, i.e., each row of data.

data.coord

A numeric vector or matrix specifying the coordinates of the observed data. If data.coord is a matrix, each row must corresponds to one location.

krig.coord

A numeric vector or matrix specifying the coordinates where the kriging predictor has to be computed. If krig.coord is a matrix, each row must correspond to one location.

cov.mod

A character string specifying the covariance function family. Must be one of "whitmat", "powexp", "cauchy", "bessel" or "caugen" for the Whittle-Matern, the powered exponential, the Cauchy, the Bessel or the generalized Cauchy covariance families.

sill, range, smooth, smooth2

Numerics specifiying the sill, range, smooth and, if any, the second smooth parameters of the covariance function.

grid

Logical. Does krig.coord specifies a grid?

only.weights

Logical. Should only the kriging weights be computed? If FALSE, the kriging predictor isn't computed.

Value

A list with components

coord

The coordinates where the kriging predictor has been computed;

krig.est

The kriging predictor estimates;

grid

Does coord define a grid?;

weights

A matrix giving the kriging weights: each column corresponds to one prediction location.

Author(s)

Mathieu Ribatet

References

Chiles, J.-P. and Delfiner, P. (1999) Geostatistics, Modeling Spatial Uncertainty Wiley Series in Probability and Statistics.

See Also

condrgp, rgp, covariance.

Examples

## Kriging from a single realisation
n.site <- 50
n.pred <- 512

x.obs <- runif(n.site, -100, 100)
x.pred <- seq(-100, 100, length = n.pred)

data <- rgp(1, x.obs, "whitmat", sill = 1, range = 10, smooth = 0.75)

krig <- kriging(data, x.obs, x.pred, "whitmat", sill = 1, range = 10,
smooth = 0.75)

plot(krig$coord, krig$krig.est, type = "l", xlab = "x", ylab =
expression(hat(Y)(x)))
points(x.obs, data, col = 2, pch = 21, bg = 2)

## Kriging from several realisations
n.real <- 3
data <- rgp(n.real, x.obs, "whitmat", sill = 1, range = 10, smooth = 0.75)

krig <- kriging(data, x.obs, x.pred, "whitmat", sill = 1, range = 10,
smooth = 0.75)

matplot(krig$coord, t(krig$krig.est), type = "l", xlab = "x", ylab =
expression(hat(Y)(x)), lty = 1)
matpoints(x.obs, t(data), pch = 21, col = 1:n.real, bg = 1:n.real)
title("Three kriging predictors in one shot")

## Two dimensional kriging on a grid
x.obs <- matrix(runif(2 * n.site, -100, 100), ncol = 2)
x <- y <- seq(-100, 100, length = 100)
x.pred <- cbind(x, y)

data <- rgp(1, x.obs, "whitmat", sill = 1, range = 10, smooth = 0.75)

krig <- kriging(data, x.obs, x.pred, "whitmat", sill = 1, range = 10,
smooth = 0.75, grid = TRUE)

z.lim <- range(c(data, krig$krig.est))
breaks <- seq(z.lim[1], z.lim[2], length = 65)
col <- heat.colors(64)
idx <- as.numeric(cut(data, breaks))

image(x, y, krig$krig.est, col = col, breaks = breaks)
points(x.obs, bg = col[idx], pch = 21)
## Note how the background colors of the above points matches the ones
## returned by the image function

Bayesian hierarchical models for spatial extremes

Description

This function generates a Markov chain from a Bayesian hierarchical model for block maxima assuming conditional independence.

Usage

latent(data, coord, cov.mod = "powexp", loc.form, scale.form,
shape.form, marg.cov = NULL, hyper, prop, start, n = 5000, thin = 1,
burn.in = 0, use.log.link = FALSE)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

cov.mod

A character string corresponding to the covariance model for the Gaussian latent processes. Must be one of "gauss" for the Smith's model; "whitmat", "cauchy", "powexp" or "bessel" or for the Whittle-Matern, the Cauchy, the Powered Exponential and the Bessel correlation families.

loc.form, scale.form, shape.form

R formulas defining the spatial linear model for the mean of the latent processes.

marg.cov

Matrix with named columns giving additional covariates for the latent processes means. If NULL, no extra covariates are used.

hyper

A named list specifying the hyper-parameters — see Details.

prop

A named list specifying the jump sizes when a Metropolis–Hastings move is needed — see Details.

start

A named list specifying the starting values — see Details.

n

The effective length of the simulated Markov chain i.e., once the burnin period has been discarded and after thinning.

thin

An integer specifying the thinning length. The default is 1, i.e., no thinning.

burn.in

An integer specifying the burnin period. The default is 0, i.e., no burnin.

use.log.link

An integer. Should a log-link function should be use for the GEV scale parameters, i.e., assuming that the GEV scale parameters a drawn from a log-normal process rather than a gaussian process.

Details

This function generates a Markov chain from the following model. For each xRdx \in R^d, suppose that Y(x)Y(x) is GEV distributed whose parameters {μ(x),σ(x),ξ(x)}\{\mu(x), \sigma(x), \xi(x)\} vary smoothly for xRdx \in R^d according to a stochastic process S(x)S(x). We assume that the processes for each GEV parameters are mutually independent Gaussian processes. For instance, we take for the location parameter μ(x)\mu(x)

μ(x)=fμ(x)(x;βμ)+Sμ(x;αμ,λμ,κμ)\mu(x) = f_{\mu(x)}(x;\beta_\mu) + S_\mu(x;\alpha_{\mu}, \lambda_\mu, \kappa_\mu)

where fμf_\mu is a deterministic function depending on regression parameters βμ\beta_\mu, and SμS_\mu is a zero mean, stationary Gaussian process with a prescribed covariance function with sill αμ\alpha_\mu, range λμ\lambda_\mu and shape parameters κμ\kappa_\mu. Similar formulations for the scale σ(x)\sigma(x) and the shape ξ(x)\xi(x) parameters are used. Then conditional on the values of the three Gaussian processes at the sites (x1,,xK)(x_1, \ldots, x_K), the maxima are assumed to follow GEV distributions

Yi(xj){μ(xj),σ(xj),ξ(xj)}GEV{μ(xj),σ(xj),ξ(xj)},Y_i(x_j) \mid \{\mu(x_j), \sigma(x_j), \xi(x_j)\} \sim \mbox{GEV}\{\mu(x_j), \sigma(x_j), \xi(x_j)\},

independently for each location (x1,,xK)(x_1, \ldots, x_K).

A joint prior density must be defined for the sills, ranges, shapes parameters of the covariance functions as well as for the regression parameters βμ\beta_\mu,βσ\beta_\sigma and βξ\beta_\xi. Conjugate priors are used whenever possible, taking independent inverse Gamma and multivariate normal distributions for the sills and the regression parameters. No conjugate prior exist for λ\lambda and κ\kappa, for wich a Gamma distribution is assumed.

Consequently hyper is a named list with named components

sills

A list with three components named 'loc', 'scale' and 'shape' each of these is a 2-length vector specifying the shape and the scale of the inverse Gamma prior distribution for the sill parameter of the covariance functions;

ranges

A list with three components named 'loc', 'scale' and 'shape' each of these is a 2-length vector specifying the shape and the scale of the Gamma prior distribution for the range parameter of the covariance functions.

smooths

A list with three components named 'loc', 'scale' and 'shape' each of these is a 2-length vector specifying the shape and the scale of the Gamma prior distribution for the shape parameter of the covariance functions;

betaMean

A list with three components named 'loc', 'scale' and 'shape' each of these is a vector specifying the mean vector of the multivariate normal prior distribution for the regression parameters;

betaIcov

A list with three components named 'loc', 'scale' and 'shape' each of these is a matrix specifying the inverse of the covariance matrix of the multivariate normal prior distribution for the regression parameters.

As no conjugate prior exists for the GEV parameters and the range and shape parameters of the covariance functions, Metropolis–Hastings steps are needed. The proposals θprop\theta_{prop} are drawn from a proposal density q(θcur,s)q(\cdot \mid \theta_{cur}, s) where θcur\theta_{cur} is the current state of the parameter and ss is a parameter of the proposal density to be defined. These proposals are driven by prop which is a list with three named components

gev

A vector of length 3 specifying the standard deviations of the proposal distributions. These are taken to be normal distribution for the location and shape GEV parameters and a log-normal distribution for the scale GEV parameters;

ranges

A vector of length 3 specifying the jump sizes for the range parameters of the covariance functions — q(θcur,s)q(\cdot | \theta_{cur}, s) is the log-normal density with mean θcur\theta_{cur} and standard deviation ss both on the log-scale;

smooths

A vector of length 3 specifying the jump sizes for the shape parameters of the covariance functions — q(θcur,s)q(\cdot | \theta_{cur}, s) is the log-normal density with mean θcur\theta_{cur} and standard deviation ss both on the log-scale.

If one want to held fixed a parameter this can be done by setting a null jump size then the parameter will be held fixed to its starting value.

Finally start must be a named list with 4 named components

sills

A vector of length 3 specifying the starting values for the sill of the covariance functions;

ranges

A vector of length 3 specifying the starting values for the range of the covariance functions;

smooths

A vector of length 3 specifying the starting values for the shape of the covariance functions;

beta

A named list with 3 components 'loc', 'scale' and 'shape' each of these is a numeric vector specifying the starting values for the regression coefficients.

Value

A list

Warning

This function can be time consuming and makes an intensive use of BLAS routines so it is (much!) faster if you have an optimized BLAS.

The starting values will never be stored in the generated Markov chain even when burn.in=0.

Note

If you want to analyze the convergence ans mixing properties of the Markov chain, it is recommended to use the library coda.

Author(s)

Mathieu Ribatet

References

Banerjee, S., Carlin, B. P., and Gelfand, A. E. (2004). Hierarchical Modeling and Analysis for Spatial Data. Chapman & Hall/CRC, New York.

Casson, E. and Coles, S. (1999) Spatial regression models for extremes. Extremes 1,449–468.

Cooley, D., Nychka, D. and Naveau, P. (2007) Bayesian spatial modelling of extreme precipitation return levels Journal of the American Statistical Association 102:479, 824–840.

Davison, A.C., Padoan, S.A. and Ribatet, M. Statistical Modelling of Spatial Extremes. Submitted.

Examples

## Not run: 
## Generate realizations from the model
n.site <- 30
n.obs <- 50
coord <- cbind(lon = runif(n.site, -10, 10), lat = runif(n.site, -10 , 10))

gp.loc <- rgp(1, coord, "powexp", sill = 4, range = 20, smooth = 1)
gp.scale <- rgp(1, coord, "powexp", sill = 0.4, range = 5, smooth = 1)
gp.shape <- rgp(1, coord, "powexp", sill = 0.01, range = 10, smooth = 1)

locs <- 26 + 0.5 * coord[,"lon"] + gp.loc
scales <- 10 + 0.2 * coord[,"lat"] + gp.scale
shapes <- 0.15 + gp.shape

data <- matrix(NA, n.obs, n.site)
for (i in 1:n.site)
  data[,i] <- rgev(n.obs, locs[i], scales[i], shapes[i])

loc.form <- y ~ lon
scale.form <- y ~ lat
shape.form <- y ~ 1

hyper <- list()
hyper$sills <- list(loc = c(1,8), scale = c(1,1), shape = c(1,0.02))
hyper$ranges <- list(loc = c(2,20), scale = c(1,5), shape = c(1, 10))
hyper$smooths <- list(loc = c(1,1/3), scale = c(1,1/3), shape = c(1, 1/3))
hyper$betaMeans <- list(loc = rep(0, 2), scale = c(9, 0), shape = 0)
hyper$betaIcov <- list(loc = solve(diag(c(400, 100))),
                       scale = solve(diag(c(400, 100))),
                       shape = solve(diag(c(10), 1, 1)))

## We will use an exponential covariance function so the jump sizes for
## the shape parameter of the covariance function are null.
prop <- list(gev = c(1.2, 0.08, 0.08), ranges = c(0.7, 0.8, 0.7), smooths = c(0,0,0))
start <- list(sills = c(4, .36, 0.009), ranges = c(24, 17, 16), smooths
              = c(1, 1, 1),  beta = list(loc = c(26, 0.5), scale = c(10, 0.2),
                               shape = c(0.15)))

mc <- latent(data, coord, loc.form = loc.form, scale.form = scale.form,
             shape.form = shape.form, hyper = hyper, prop = prop, start = start,
             n = 10000, burn.in = 5000, thin = 15)
mc

## End(Not run)

Computes the lambda-madogram

Description

Computes the lambda-madogram for max-stable processes.

Usage

lmadogram(data, coord, n.bins, xlab, ylab, zlab, n.lambda = 11, marge =
"emp", col = terrain.colors(50, alpha = 0.5), theta = 90, phi = 20,
border = NA, ...)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

n.bins

The number of bins to be used. If missing, pairwise lambda-madogram estimates will be computed.

xlab, ylab, zlab

The x-axis, y-axis and z-axis labels. May be missing.

n.lambda

Integer giving the number of lambda values.

marge

Character string. If 'emp', probabilities of non exceedances are estimated using the empirical CDF. If 'mle' (default), maximum likelihood estimates are used.

col

The colors used to emphasize the gradient of the lambda-madogram.

theta, phi, border

Options to be passed to the persp function.

...

Additional options to be passed to the persp function.

Details

Let Z(x)Z(x) be a stationary process. The λ\lambda-madogram is defined as follows:

νλ(h)=12E[Fλ(Z(x+h))F1λ(Z(x))]\nu_{\lambda}(h) = \frac{1}{2}\mbox{E}\left[|F^\lambda(Z(x+h)) - F^{1-\lambda}(Z(x))| \right]

Value

A graphic and (invisibly) a matrix with the lag distances, the λ\lambda-madogram estimate.

Author(s)

Mathieu Ribatet

References

Naveau, P., Guillou, A., Cooley, D. and Diebolt, J. (2009) Modelling Pairwise Dependence of Maxima in Space. To appear in Biometrika.

See Also

madogram, fmadogram

Examples

n.site <- 50
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(40, locations, cov.mod = "whitmat", nugget = 0, range = 1,
smooth = 2)

##Compute the lambda-madogram
lmadogram(data, locations, n.bins = 80)

Extracts Log-Likelihood

Description

Extract the pairwise log-likelihood for objects of class “maxstab” and “copula”

Usage

## S3 method for class 'maxstab'
logLik(object, ...)
## S3 method for class 'copula'
logLik(object, ...)

Arguments

object

An object of class “maxstab” or “copula”. Most often this will be the output of the fitmaxstab of fitcopula functions.

...

Other arguments to be passed to the logLik function.

Value

Standard logLik object (see logLik) except that it might actually correspond to the pairwise log-likelihood, e.g., for the class “maxstab”!

Author(s)

Mathieu Ribatet

See Also

logLik

Examples

##Define the coordinates of each location
n.site <- 30
locations <- matrix(5 + runif(2*n.site, 0, 10), ncol = 2)

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(30, locations, cov.mod = "whitmat", nugget = 0, range = 3,
smooth = 0.5)
fit <- fitmaxstab(data, locations, "whitmat")
logLik(fit)

Estimates the spatial dependence parameter of a max-stable process by minimizing least squares.

Description

Estimates the spatial dependence parameter of a max-stable process by minimizing least squares.

Usage

lsmaxstab(data, coord, cov.mod = "gauss", marge = "emp", control =
list(), iso = FALSE, ..., weighted = TRUE)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

cov.mod

Character string specifying the max-stable process considered. Must be one of "gauss" (Smith's model), "whitmat", "cauchy", "powexp", "bessel", "caugen" for the Schlather model with the corresponding correlation function.

marge

Character string specifying how margins are transformed to unit Frechet. Must be one of "emp", "frech" or "mle" - see function fitextcoeff.

control

The control arguments to be passed to the optim function.

iso

Logical. If TRUE, isotropy is supposed. Otherwise (default), anisotropy is allowed. Currently this is only useful for the Smith model.

...

Optional arguments.

weighted

Logical. Should weighted least squares be used? See Details.

Details

The fitting procedure is based on weighted least squares. More precisely, the fitting criteria is to minimize:

i,j(θ~i,jθ^i,jsi,j)2\sum_{i,j} \left(\frac{\tilde{\theta}_{i,j} - \hat{\theta}_{i,j}}{s_{i,j}}\right)^2

where θ~i,j\tilde{\theta}_{i,j} is a non parametric estimate of the extremal coefficient related to location i and j, θ^i,j\hat{\theta}_{i,j} is the fitted extremal coefficient derived from the maxstable model considered and si,js_{i,j} are the standard errors related to the estimates θ~i,j\tilde{\theta}_{i,j}.

Value

An object of class maxstab.

Author(s)

Mathieu Ribatet

References

Smith, R. L. (1990) Max-stable processes and spatial extremes. Unpublished manuscript.

See Also

fitcovariance, fitmaxstab, fitextcoeff

Examples

n.site <- 50
n.obs <- 100
locations <- matrix(runif(2*n.site, 0, 40), ncol = 2)
colnames(locations) <- c("lon", "lat")

## Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(50, locations, cov.mod = "gauss", cov11 = 200, cov12 =
0, cov22 = 200)

lsmaxstab(data, locations, "gauss")

##Force an isotropic model and do not use weights
lsmaxstab(data, locations, "gauss", iso = TRUE, weighted = FALSE)

Computes madograms

Description

Computes the madogram for max-stable processes.

Usage

madogram(data, coord, fitted, n.bins, gev.param = c(0, 1, 0), which =
c("mado", "ext"), xlab, ylab, col = c(1, 2), angles = NULL, marge =
"emp", add = FALSE, xlim = c(0, max(dist)), ...)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

fitted

An object of class maxstab - usually the output of the fitmaxstab function. May be missing.

n.bins

The number of bins to be used. If missing, pairwise madogram estimates will be computed.

gev.param

Numeric vector of length 3 specifying the location, scale and shape parameters for the GEV.

which

A character vector of maximum size 2. It specifies if the madogram and/or the extremal coefficient functions have to be plotted.

xlab, ylab

The x-axis and y-axis labels. May be missing. Note that ylab must have the same length has which.

col

The colors used for the points and optionnaly for the fitted curve.

angles

A numeric vector. A partition of the interval (0,π)(0, \pi) to help detecting anisotropy.

marge

Character string. If 'emp', the observation are first transformed to the unit Frechet scale by using the empirical CDF. If 'mle' (default), maximum likelihood estimates are used.

add

Logical. If TRUE, the plot is added to the current figure; otherwhise (default) a new plot is computed.

xlim

A numeric vector of length 2 specifying the x coordinate range.

...

Additional options to be passed to the plot function.

Details

Let Z(x)Z(x) be a stationary process. The madogram is defined as follows:

ν(h)=12E[Z(x+h)Z(x)]\nu(h) = \frac{1}{2}\mbox{E}\left[|Z(x+h) - Z(x)| \right]

If now Z(x)Z(x) is a stationary max-stable random field with GEV marginals. Provided the GEV shape parameter ξ\xi is such that ξ<1\xi < 1. The extremal coefficient θ(h)\theta(h) satisfies:

θ(h)={uβ(μ+ν(h)Γ(1ξ)),ξ0exp(ν(h)σ),ξ=0\theta(h) = \left\{ \begin{array}{ll} u_\beta \left(\mu + \frac{\nu(h)}{\Gamma(1 - \xi)} \right), &\xi \neq 0\\ \exp\left(\frac{\nu(h)}{\sigma}\right), &\xi = 0 \end{array} \right.

where Γ\Gamma is the gamma function and uβu_\beta is defined as follows:

uβ(u)=(1+ξuμσ)+1/ξu_\beta(u) = \left(1 + \xi \frac{u - \mu}{\sigma} \right)_+^{1/\xi}

and β=(μ,σ,ξ)\beta = (\mu, \sigma, \xi), i.e, the vector of the GEV parameters.

Value

A graphic and (invisibly) a matrix with the lag distances, the madogram and extremal coefficient estimates.

Author(s)

Mathieu Ribatet

References

Cooley, D., Naveau, P. and Poncet, P. (2006) Variograms for spatial max-stable random fields. Dependence in Probability and Statistics, 373–390.

See Also

fmadogram, lmadogram

Examples

n.site <- 15
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(40, locations, cov.mod = "whitmat", nugget = 0, range = 1,
smooth = 2)

##Compute the madogram
madogram(data, locations)

##Compare the madogram with a fitted max-stable model
fitted <- fitmaxstab(data, locations, "whitmat", nugget = 0)
madogram(fitted = fitted, which = "ext")

Produces a 2D map from a fitted max-stable process

Description

Produces a 2D map from a fitted max-stable process.

Usage

map(fitted, x, y, covariates = NULL, param = "quant", ret.per = 100, col
= terrain.colors(64), plot.contour = TRUE, ...)

Arguments

fitted

An object of class maxstab. Most often, it will be the output of the function fitmaxstab.

x, y

Numeric vector that gives the coordinates of the grid.

covariates

An array specifying the covariates at each grid point defined by x and y. If NULL, no covariate is needed. See the example to see how to build it.

param

A character string. Must be one of "loc", "scale", "shape" or "quant" for a map of the location, scale, shape parameters or for a map of a specified quantile.

ret.per

A numeric giving the return period for which the quantile map is plotted. It is only required if param = "quant".

col

A list of colors such as that generated by 'rainbow', 'heat.colors', 'topo.colors', 'terrain.colors' or similar functions.

plot.contour

Logical. If TRUE (default), contour lines are added to the plot.

...

Several arguments to be passed to the image function.

Value

A plot. Additionally, a list with the details for plotting the map is returned invisibly.

Author(s)

Mathieu Ribatet

See Also

condmap, filled.contour, heatmap, heat.colors, topo.colors, terrain.colors, rainbow

Examples

##We run an artifical example using the volcano data set as a study
##region
dim <- dim(volcano)
n.x <- dim[1]
n.y <- dim[2]

x <- 10 * 1:n.x
y <- 10 * 1:n.y

n.site <- 15
idx.x <- sample(n.x, n.site)
idx.y <- sample(n.y, n.site)
locations <- cbind(lon = x[idx.x], lat = y[idx.y])
alt <- diag(volcano[idx.x, idx.y])

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(40, locations, cov.mod = "whitmat", nugget = 0, range = 750,
smooth = 1)

##Now define the spatial model for the GEV parameters
param.loc <- -10 - 0.04 * locations[,1] + alt / 5
param.scale <- 5 - locations[,2] / 30 + alt / 4
param.shape <- rep(.2, n.site)

##Transform the unit Frechet margins to GEV
for (i in 1:n.site)
  data[,i] <- frech2gev(data[,i], param.loc[i], param.scale[i],
param.shape[i])

##Define a model for the GEV margins to be fitted
##shape ~ 1 stands for the GEV shape parameter is constant
##over the region
loc.form <- loc ~ lon + alt
scale.form <- scale ~ lat + alt
shape.form <- shape ~ 1

##  1- Fit a max-stable process
schlather <- fitmaxstab(data, locations, "whitmat", loc.form, scale.form,
                        shape.form, marg.cov = cbind(alt = alt), nugget
= 0)

##  2- Produce a map of the pointwise 50-year return level

##Here we have only one covariate i.e. alt
n.cov <- 1
covariates <- array(volcano, dim = c(n.x, n.y, n.cov), dimnames =
list(NULL, NULL, "alt"))

par(mfrow = c(1,2))
image(x, y, volcano, col = terrain.colors(64), main = "Elevation map")
map(schlather, x, y, covariates, ret.per = 50, plot.contour = FALSE,
main = "50-year return level")

Two dimensional map from a Bayesian hierarchical model

Description

This function plots 2D maps from a Markov chain.

Usage

map.latent(fitted, x, y, covariates = NULL, param = "quant", ret.per =
100, col = terrain.colors(64), plot.contour = TRUE, fun = mean, level =
0.95, show.data = TRUE, control = list(nlines = 500), ...)

Arguments

fitted

An object of class "latent". Typically this will be the output of latent.

x, y

Numeric vector specifying the coordinates of the grid points.

covariates

An array specifying the covariates at each grid point defined by x and y. If NULL, no covariate is needed.

param

A character string. Must be one of "loc", "scale", "shape" or "quant" for a map of the location, scale, shape parameters or for a map of a specified quantile.

ret.per

A numeric giving the return period for which the quantile map is plotted. It is only required if param = "quant".

col

A list of colors such as that generated by 'rainbow', 'heat.colors', 'topo.colors', 'terrain.colors' or similar functions.

plot.contour

Logical. If TRUE (default), contour lines are added to the plot.

fun

A character string specifying the function to be used to get posterior point estimates. The default is to take posterior means.

level

A numeric specifying the significance level for the pointwise credible intervals.

show.data

Logical. Should the locations where have observed the process have to be plotted?

control

A list with named components specifying options to be passed to rgp. Typically one might want specify nlines to reduce the computational demand.

...

Several arguments to be passed to the image function.

Value

A plot and a invisible list containing all the data required to do the plot.

Author(s)

Mathieu Ribatet

See Also

condrgp, map

Examples

## Not run: 
## Generate realizations from the model
n.site <- 30
n.obs <- 50
coord <- cbind(lon = runif(n.site, -10, 10), lat = runif(n.site, -10 , 10))

gp.loc <- rgp(1, coord, "powexp", sill = 4, range = 20, smooth = 1)
gp.scale <- rgp(1, coord, "powexp", sill = 0.4, range = 5, smooth = 1)
gp.shape <- rgp(1, coord, "powexp", sill = 0.01, range = 10, smooth = 1)

locs <- 26 + 0.5 * coord[,"lon"] + gp.loc
scales <- 10 + 0.2 * coord[,"lat"] + gp.scale
shapes <- 0.15 + gp.shape

data <- matrix(NA, n.obs, n.site)
for (i in 1:n.site)
  data[,i] <- rgev(n.obs, locs[i], scales[i], shapes[i])

loc.form <- y ~ lon
scale.form <- y ~ lat
shape.form <- y ~ 1

hyper <- list()
hyper$sills <- list(loc = c(1,8), scale = c(1,1), shape = c(1,0.02))
hyper$ranges <- list(loc = c(2,20), scale = c(1,5), shape = c(1, 10))
hyper$smooths <- list(loc = c(1,1/3), scale = c(1,1/3), shape = c(1, 1/3))
hyper$betaMeans <- list(loc = rep(0, 2), scale = c(9, 0), shape = 0)
hyper$betaIcov <- list(loc = solve(diag(c(400, 100))),
                       scale = solve(diag(c(400, 100))),
                       shape = solve(diag(c(10), 1, 1)))

## We will use an exponential covariance function so the jump sizes for
## the shape parameter of the covariance function are null.
prop <- list(gev = c(2.5, 1.5, 0.2), ranges = c(0.7, 0.75, 0.9), smooths = c(0,0,0))
start <- list(sills = c(4, .36, 0.009), ranges = c(24, 17, 16), smooths
              = c(1, 1, 1),  beta = list(loc = c(26, 0.5), scale = c(10, 0.2),
                               shape = c(0.15)))

## Generate a Markov chain
mc <- latent(data, coord, loc.form = loc.form, scale.form = scale.form,
             shape.form = shape.form, hyper = hyper, prop = prop, start = start,
             n = 100)

x.grid <- y.grid <- seq(-10, 10, length = 50)
map.latent(mc, x.grid, y.grid, param = "shape")

## End(Not run)

Fits univariate extreme value distributions to data

Description

These functions fit the generalised extreme value and generalised Pareto distribution to data using maximum likelihood.

Usage

gevmle(x, ..., method = "Nelder")
gpdmle(x, threshold, ..., method = "Nelder")

Arguments

x

Numeric vector of observations

...

Optional arguments to be passed to the optim function.

threshold

Numeric. The threshold value.

method

The numerical optimisation method to be used.

Details

These two functions are “extremely light” functions to fit the GEV/GPD. These functions are mainly useful to compute starting values for the Schlather and Smith model - see fitmaxstab.

If more refined (univariate) analysis have to be performed, users should use more specialised packages - e.g. POT, evd, ismev, ....

Value

A vector for the estimated parameters of the GEV/GPD.

Author(s)

Mathieu Ribatet

Examples

## 1 - GEV fit
x <- rep(NA, 100)
for (i in 1:100)
  x[i] <- max(rnorm(365))

gevmle(x)

## 2- GPD fit
x <- rnorm(10000)
##we need to fix a threshold
u <- quantile(x, 0.99)
gpdmle(x, u)

Define a model for the spatial behaviour of the GEV parameters

Description

This function defines the model for the spatial behaviour of the GEV parameter.

Usage

modeldef(data, formula)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

formula

A R formula. See details for further details.

Value

need to be documented

Author(s)

Mathieu Ribatet

See Also

formula

Examples

## 1- A design matrix from a classical linear model 
n.site <- 5
coord <- matrix(rnorm(2*n.site, sd = sqrt(.2)), ncol = 2)
colnames(coord) <- c("lon", "lat")
loc.form <- loc ~ lat + I(lon^2)
modeldef(coord, loc.form)

## 2- A design and penalization matrix from a penalized smoothin spline
x <- sort(runif(10, -2, 10))
n.knots <- 3
knots <- quantile(x, prob = 1:n.knots / (n.knots + 2))
modeldef(x, y ~ rb(x, knots = knots, degree = 3, penalty = 1))

Model checking of a fitted copula based model.

Description

This function produces several plots to assess the goodness of fit of a fitted copula based model for spatial extremes.

Usage

## S3 method for class 'copula'
plot(x, ..., sites)

Arguments

x

An object of class copula. Most often, this will be the output of fitcopula.

...

Here for compatibility reasons but not yet implemented.

sites

A vector of integer of length 4 specifying the locations to be used for the model checking. If missing, locations will be choosen randomly.

Details

The diagonal plots are return level plots. The lower ones are qq-plots (on the Gumbel scale) between observed pairwise maxima for each block, e.g. year, and the ones obtained by simulations from the fitted model. The upper plot compares the fitted extremal coefficient functions to semi-empirical estimates from the F-madogram - see fmadogram. The two remaining plots are the stations locations and a qq-plot of blockwise maxima where the block size is 4.

Value

Several diagnostic plots.

Author(s)

Mathieu Ribatet

Examples

## Not run: 
n.site <- 20
n.obs <- 50
coord <- matrix(runif(2 * n.site, 0, 10), ncol = 2)
colnames(coord) <- c("lon", "lat")
data <- rmaxstab(n.obs, coord, "powexp", nugget = 0, range = 3, smooth =
1)
fitted <- fitcopula(log(data), coord, "student", "powexp", y ~ 1, y ~ 1, y ~ 1,
nugget = 0)
plot(fitted)

## End(Not run)

Model checking of a fitted max-stable model

Description

This function produces several plots to assess the goodness of fit of a fitted max-stable model.

Usage

## S3 method for class 'maxstab'
plot(x, ..., sites)

Arguments

x

An object of class maxstab. Most often, this will be the output of fitmaxstab or lsmaxstab.

...

Here for compatibility reasons but not yet implemented.

sites

A vector of integer of length 4 specifying the locations to be used for the model checking. If missing, locations will be choosen randomly.

Details

The diagonal plots are return level plots. The lower ones are qq-plots (on the Gumbel scale) between observed pairwise maxima for each block, e.g. year, and the ones obtained by simulations from the fitted model. The upper plot compares the fitted extremal coefficient functions to semi-empirical estimates from the F-madogram - see fmadogram. The two remaining plots are the stations locations and a qq-plot of blockwise maxima where the block size is 4.

Value

Several diagnostic plots.

Author(s)

Mathieu Ribatet

Examples

n.site <- 20
n.obs <- 50
coord <- matrix(runif(2 * n.site, 0, 10), ncol = 2)
colnames(coord) <- c("lon", "lat")
data <- rmaxstab(n.obs, coord, "powexp", nugget = 0, range = 3, smooth =
1)
fitted <- fitmaxstab(log(data), coord, "powexp", y ~ 1, y ~ 1, y ~ 1,
nugget = 0)
plot(fitted)

Prediction of the marginal parameters for various models

Description

This function predicts the marginal GEV parameters from a fitted max-stable process, copula, penalized spline or spatial GEV model.

Usage

## S3 method for class 'maxstab'
predict(object, newdata, ret.per = NULL, std.err =
TRUE, ...)
## S3 method for class 'copula'
predict(object, newdata, ret.per = NULL, std.err =
TRUE, ...)
## S3 method for class 'pspline2'
predict(object, newdata, ...)
## S3 method for class 'spatgev'
predict(object, newdata, ret.per = NULL, ...)

Arguments

object

An object of class 'maxstab', 'copula', 'pspline' or 'spatgev'. Most often, it will be the output of one of the following functions: rbpspline, fitmaxstab, fitcopula or fitspatgev.

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.

std.err

Logical. If TRUE (default), standard erros will be computed if possible.

...

further arguments passed to or from other methods.

Value

Print several information on screen.

Author(s)

Mathieu Ribatet

Examples

## 1- Simulate a max-stable random field
n.site <- 35
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

data <- rmaxstab(50, locations, cov.mod = "whitmat", nugget = 0, range = 30,
smooth = 0.5)

## 2- Transformation to non unit Frechet margins
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)
  data[,i] <- frech2gev(data[,i], param.loc[i], param.scale[i],
param.shape[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(data, 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)

Printing objects of classes defined in the SpatialExtreme packages

Description

Methods for printing objects of classes introduced by the SpatialExtremes package.

Usage

## S3 method for class 'pspline2'
print(x, ...)
## S3 method for class 'maxstab'
print(x, digits = max(3, getOption("digits") - 3), ...)
## S3 method for class 'copula'
print(x, digits = max(3, getOption("digits") - 3), ...)
## S3 method for class 'spatgev'
print(x, digits = max(3, getOption("digits") - 3), ...)
## S3 method for class 'latent'
print(x, digits = max(3, getOption("digits") - 3), ...,
 level = 0.95)

Arguments

x

An object of class 'pspline', 'maxstab', 'copula', 'spatgev' or 'latent'. Most often, x will be the output of one of the following functions: rbpspline, fitmaxstab, fitcopula, fitspatgev or latent.

digits

The number of digits to be printed.

...

Other options to be passed to the print function.

level

A numeric giving the significance level for the credible intervals—class 'latent' only.

Value

Print several information on screen.

Author(s)

Mathieu Ribatet

Examples

##Define the coordinates of each location
n.site <- 30
coord <- matrix(5 + rnorm(2*n.site, sd = sqrt(2)), ncol = 2)
colnames(coord) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(30, coord, cov.mod = "whitmat", nugget = 0, range = 3,
smooth = 0.5)

## Printing max-stable objects
fit <- fitmaxstab(data, coord, "whitmat")
fit

## Printing spatial GEV objects
loc.form <- scale.form <- shape.form <- y ~ 1
fit <- fitspatgev(data, coord, loc.form, scale.form, shape.form)
fit

Method for profiling fitted max-stable objects

Description

Computes profile traces for fitted max-stable models.

Usage

## S3 method for class 'maxstab'
profile(fitted, param, range, n = 10, plot = TRUE,
conf = 0.90, method = "RJ", square = "chol", ...)

Arguments

fitted

An object of class “maxstab”. Most often, it will be the output of the function fitmaxstab.

param

A character string giving the model parameter that are to be profiled.

range

The range for the profiled model parameter that must be explored.

n

Integer. The number of profiled model parameter that must be considered.

plot

Logical. If TRUE (default), the profile trace is plotted.

conf

Numeric giving the confidence interval level.

method

Character string. Must be one of "CB", "RJ" or "none" for the Chandler and Bate or the Rotnitzky and Jewell approaches respectively. The "none" method simply plots the profile of the log-composite likelihood. See details.

square

The choice for the matrix square root. This is only useful for the 'CB' method. Must be one of 'chol' (Cholesky) or 'svd' (Singular Value Decomposition).

...

Extra options that must be passed to the plot function.

Details

The Rotnitzky and Jewell approach consists in adjusting the distribution of the likelihood ratio statistics - which under misspecification is no longer χ2\chi^2 distributed.

The Chandler and Bate approach adjusts the composite likelihood itself is such a way that the usual asymptotic χ2\chi^2 null distribution is preserved. Note that in the current code, we use the singular value decomposition for the computation of matrix square roots to preserve asymmetry in the profile composite likelihood.

Value

A matrix. The first column corresponds to the values for which the profiled model parameter is fixed. The second column gives the value of the pairwise log-likelihood. The remaining columns contain the constrained maximum likelihood estimates for the remaining model parameters.

Warnings

This function can be really time consuming!

Author(s)

Mathieu Ribatet

References

Chandler, R. E. and Bate, S. (2007) Inference for clustered data using the independence loglikelihood Biometrika, 94, 167–183.

Rotnitzky, A. and Jewell, N. (1990) Hypothesis testing of regression parameters in semiparametric generalized linear models for cluster correlated data. Biometrika 77, 485–97.

Examples

## Not run: 
##Define the coordinates of each location
n.site <- 30
locations <- matrix(rnorm(2*n.site, sd = sqrt(.2)), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(50, locations, cov.mod = "gauss", cov11 = 100, cov12 =
25, cov22 = 220)

##Fit a max-stable process
##  1- using the Smith's model
fitted <- fitmaxstab(data, locations, "gauss", fit.marge = FALSE)

##Plot the profile pairwise log-likelihood for the ''cov11'' parameter
profile(fitted, "cov11", range = c(20, 180))

## End(Not run)

Method for profiling (in 2d) fitted max-stable objects

Description

Computes profile surfaces for fitted max-stable models.

Usage

## S3 method for class 'maxstab'
profile2d(fitted, params, ranges, n = 10, plot = TRUE,
...)

Arguments

fitted

An object of class “maxstab”. Most often, it will be the output of the function fitmaxstab.

params

A character vector giving the two model parameters that are to be profiled.

ranges

A matrix corresponding to the ranges for the profiled model parameters that must be explored. Each row corresponds to one model parameter range.

n

Integer. The number of profiled model parameter that must be considered.

plot

Logical. If TRUE (default), the profile surface is plotted.

...

Extra options that must be passed to the plot function.

Value

A list with two arguments: coord and llik. coord is a matrix representing the grid where the profiled model parameters are fixed. llik the corresponding pairwise log-likelihood.

Warnings

This function can be really time consuming!

Author(s)

Mathieu Ribatet

Examples

## Not run: 
##Define the coordinates of each location
n.site <- 30
locations <- matrix(rnorm(2*n.site, sd = sqrt(.2)), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(30, locations, cov.mod = "whitmat", nugget = 0, range = 30,
smooth = 0.5)

##Now define the spatial model for the GEV parameters
param.loc <- -10 + 2 * locations[,2]
param.scale <- 5 + 2 * locations[,1] + locations[,2]^2
param.shape <- rep(0.2, n.site)

##Transform the unit Frechet margins to GEV
for (i in 1:n.site)
  data[,i] <- frech2gev(data[,i], param.loc[i], param.scale[i],
param.shape[i]) 

##Define a model for the GEV margins to be fitted
##shape ~ 1 stands for the GEV shape parameter is constant
##over the region
loc.form <- loc ~ lat
scale.form <- scale ~ lon + (lat^2)
shape.form <- shape ~ 1

##Fit a max-stable process
##  1- using the Schlather representation
fitted <- fitmaxstab(data, locations, "whitmat", loc.form, scale.form,
                     shape.form)

##Plot the profile pairwise log-likelihood for the smooth parameter
ranges <- rbind(c(9,11), c(.3, .8))
profile2d(fitted, c("range", "smooth"), ranges = ranges)

## End(Not run)

QQ-plot for the extremal coefficient

Description

This function compares the extremal coefficients estimated from a fitted max-stable process to the ones estimated semi-parametrically.

Usage

qqextcoeff(fitted, estim = "ST", marge = "emp", xlab = "Semi-Empirical",
ylab = "Model", ...)

Arguments

fitted

An object of class maxstab. Most often, this will be the output of fitmaxstab, fitcovmat or fitcovariance.

estim, marge

The estim and marge options to be passed to the fitextcoeff function.

xlab, ylab

The x and y-axis labels.

...

Optional arguments to be passed to the plot function.

Value

A QQ-plot.

Author(s)

Mathieu Ribatet

References

Schlather, M. (2002) Models for Stationary Max-Stable Random Fields. Extremes 5:1, 33–44.

Schlather, M. and Tawn, J. A. (2003) A dependence measure for multivariate and spatial extreme values: Properties and inference. Biometrika 90(1):139–156.

Smith, R. L. (1990) Max-stable processes and spatial extremes. Unpublished manuscript.

See Also

fitmaxstab, fitextcoeff

Examples

##Define the coordinate of each location
n.site <- 30
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(50, locations, cov.mod = "gauss", cov11 = 10, cov12 =
5, cov22 = 22)

fitted <- fitmaxstab(data, locations, "gauss")
qqextcoeff(fitted)

QQ-plot for the GEV parameters

Description

This function compares the GEV parameters estimated separately for each location to the ones estimated from a fitted spatial model.

Usage

qqgev(fitted, xlab, ylab, ...)

Arguments

fitted

An object of class maxstab or spatgev. Most often, this will be the output of fitmaxstab, fitcovmat, fitcovariance or fitspatgev.

xlab, ylab

The x and y-axis labels. May be missing.

...

Optional arguments to be passed to the plot function.

Value

A QQ-plot.

Author(s)

Mathieu Ribatet

References

Schlather, M. (2002) Models for Stationary Max-Stable Random Fields. Extremes 5:1, 33–44.

Schlather, M. and Tawn, J. A. (2003) A dependence measure for multivariate and spatial extreme values: Properties and inference. Biometrika 90(1):139–156.

Smith, R. L. (1990) Max-stable processes and spatial extremes. Unpublished manuscript.

See Also

qqextcoeff

Examples

##Define the coordinate of each location
n.site <- 30
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(50, locations, cov.mod = "gauss", cov11 = 100, cov12 =
25, cov22 = 220)

##Now define the spatial model for the GEV parameters
param.loc <- -10 + 2 * locations[,2]
param.scale <- 5 + 2 * locations[,1] + locations[,2]^2
param.shape <- rep(0.2, n.site)

##Transform the unit Frechet margins to GEV 
for (i in 1:n.site)
  data[,i] <- frech2gev(data[,i], param.loc[i], param.scale[i], param.shape[i])

##Define a model for the GEV margins to be fitted
##shape ~ 1 stands for the GEV shape parameter is constant
##over the region
loc.form <- loc ~ lat
scale.form <- scale ~ lon + I(lat^2)
shape.form <- shape ~ 1

fitted <- fitspatgev(data, locations, loc.form = loc.form, scale.form =
scale.form, shape.form = shape.form)

qqgev(fitted)

Summer annual maxima precipitation in Switzerland

Description

Maximum daily rainfall amounts over the years 1962–2008 occuring during June–August at 79 sites in Switzerland.

Usage

data(rainfall)

Format

This data set contains two R objects: 'rain' and 'coord'. 'rain' is a 47 by 79 matrix giving the amount of rain in millimeters, each column correspond to one locations. 'coord' is a 79 by 3 matrix giving the longitude, latitude and the elevation for each station, all of them being in meters.

Author(s)

Mathieu Ribatet

Examples

data(rainfall)
op <- par(mfrow = c(1,2),pty = "s", mar = c(0,0,0,0))
swiss(city = TRUE)
idx.site <- c(1, 10, 20)
points(coord[-idx.site,])
points(coord[idx.site,], pch = 15, col = 2:4)

par(mar = c(2,4,0,0))
plot(1962:2008, rain[,1], type = "b", xlab = "Year", ylab =
"Precipitation (cm)", ylim = c(0, 120), col = 2)
lines(1962:2008, rain[,10], col = 3, type = "b")
lines(1962:2008, rain[,20], col = 4, type = "b")
par(op)

Creates a model using penalized smoothing splines

Description

Creates a model using penalized smoothing splines using radial basis functions

Usage

rb(..., knots, degree, penalty)

Arguments

...

The explicative variables for which the spline is based on.

knots

The coordinates of knots. See section details.

degree

Numeric. The degree of the spline.

penalty

Numeric. The penalty coefficient.

Details

If one explicative variable is given in "...", the knots should be a numeric vector. Otherwise, knots should be a matrix with the same number of column and covariates.

Value

A list giving all the required information to fit a penalized smoothing spline:

dsgn.mat

The design matrix.

pen.mat

The penalization matrix.

degree

The degree of the smoothing spline.

penalty

The penalty of the smoothing spline.

knots

The knots of the smoothing spline.

data

The explicative variables (e.g. covariates).

call

How was the rb function was called?

Warning

This function is not supposed to be called directly. rb is supposed to be embedded in a R formula.

Author(s)

Mathieu Ribatet

See Also

fitmaxstab

Examples

n.site <- 30
locations <- matrix(runif(2*n.site, 0, 10), ncol = 2)
colnames(locations) <- c("lon", "lat")
knots <- quantile(locations[,2], 1:5/6)

form <- y ~ rb(lat, knots = knots, degree = 3, penalty = .5)

Fits a penalized spline with radial basis functions to data

Description

Fits a penalized spline with radial basis functions to data.

Usage

rbpspline(y, x, knots, degree, penalty = "gcv", ...)

Arguments

y

The response vector.

x

A vector/matrix giving the values of the predictor variable(s). If x is a matrix, each row corresponds to one observation.

knots

A vector givint the coordinates of the knots.

degree

The degree of the penalized smoothing spline.

penalty

A numeric giving the penalty coefficient for the penalization term. Alternatively, it could be either 'cv' or 'gcv' to choose the penalty using the (generalized) cross-validation criterion.

...

Additional options to be passed to the cv or gcv function.

Details

The penalized spline with radial basis is defined by:

f(x)=β0+β1x++βm1xm1+k=0K1βm+kxκk2m1f(x) = \beta_0 + \beta_1 x + \ldots + \beta_{m-1} x^{m-1} + \sum_{k=0}^{K-1} \beta_{m+k} || x - \kappa_k ||^{2m - 1}

where βi\beta_i are the coefficients to be estimated, κi\kappa_i are the coordinates of the i-th knot and m=d+12m = \frac{d+1}{2} where dd corresponds to the degree of the spline.

The fitting criterion is to minimize

yXβ2+λ2m1βTKβ||y - X\beta||^2 + \lambda^{2m-1} \beta^T K \beta

where λ\lambda is the penalty coefficient and KK the penalty matrix.

Value

An object of class pspline.

Author(s)

Mathieu Ribatet

References

Ruppert, D. Wand, M.P. and Carrol, R.J. (2003) Semiparametric Regression Cambridge Series in Statistical and Probabilistic Mathematics.

See Also

cv, gcv

Examples

n <- 200
x <- runif(n)
fun <- function(x) sin(3 * pi * x)
y <- fun(x) + rnorm(n, 0, sqrt(0.4))
knots <- quantile(x, prob = 1:(n/4) / (n/4 + 1))
fitted <- rbpspline(y, x, knots = knots, degree = 3)
fitted

plot(x, y)
lines(fitted, col = 2)

Simulation from copula based models with unit Frechet margins

Description

This function generates realisations from the Gaussian and Student copula with unit Frechet margins.

Usage

rcopula(n, coord, copula = "gaussian", cov.mod = "whitmat", grid =
FALSE, control = list(), nugget = 0, range = 1, smooth = 1, DoF = 1)

Arguments

n

Integer. The number of observations.

coord

A vector or matrix that gives the coordinates of each location. Each row corresponds to one location - if any

copula

A character string that specifies the copula to be used, i.e., "gaussian" or "student".

cov.mod

A character string that gives the correlation function family to be used. This must be one of "whitmat", "cauchy", "powexp" and "bessel" for the Whittle-Matern, the cauchy, the powered exponential and the bessel correlation functions.

grid

Logical. Does the coordinates represent grid points?

control

A named list that control the simulation of the gaussian process — see rgp.

nugget, range, smooth, DoF

Numerics. The parameters of the copula.

Value

A matrix containing observations from the required max-stable model. Each column represents one stations. If grid = TRUE, the function returns an array of dimension nrow(coord) x nrow(coord) x n.

Author(s)

Mathieu Ribatet

References

Demarta, S. and McNeil, A. J. (2005) The t Copula and Related Copulas International Statistical Review 73:1, 111–129.

Davison, A. C., Padoan, S. A. and Ribatet, M. (2010) Statistical Modelling of Spatial Extremes Submitted to Statistical Science.

See Also

fitcopula, rmaxstab, rmaxlin

Examples

n.site <- 25
n.obs <- 50

coord <- matrix(runif(2 * n.site, 0, 10), ncol = 2)

data1 <- rcopula(n.obs, coord, "student", "whitmat", range = 3, DoF = 3)

x <- y <- seq(0, 10, length = 100)
data2 <- rcopula(1, cbind(x, y), "gaussian", "whitmat", range = 3, grid
= TRUE)
image(x, y, log(data2), col = rainbow(64))

Gaussian Random Fields Simulation

Description

This functions generates gaussian random fields.

Usage

rgp(n, coord, cov.mod = "powexp", mean = 0, nugget = 0, sill = 1, range
= 1, smooth = 1, grid = FALSE, control = list())

Arguments

n

Integer. The number of replications.

coord

The locations coordinates for which the gaussian process is observed.

cov.mod

Character string. The covariance model used. Must be one of "whitmat", "cauchy", "powexp" of "cauchy". See the covariance function.

mean

Numeric. The mean of the gaussian random field.

nugget

Numeric. The nugget of the gaussian random field.

sill

Numeric. The sill parameter in the covariance function.

range

Numeric. The range parameter in the covariance function.

smooth

Numeric. The smooth parameter in the covariance function.

grid

Logical. Does coord defines a grid?

control

A named list with arguments 'nlines' (number of lines of the TBM simulation) and 'method' the name of the simulation method - must be one of 'exact', 'tbm' or 'circ'. If 'method' is NULL (default), the function tries to find the most appropriate simulation technique. If 'nlines' is NULL it is set to 1000.

Value

A matrix or an array containing the random field replicates.

Author(s)

Mathieu Ribatet

See Also

link{rmaxstab}

Examples

x <- y <- seq(0, 20, length = 100)
coord <- cbind(x, y)
gp <- rgp(1, coord, cov.mod = "whitmat", grid = TRUE)
filled.contour(x, y, gp, color.palette = terrain.colors)

Simulation from max-linear models

Description

This function generates realisations from a max-linear model.

Usage

rmaxlin(n, coord, cov.mod = "gauss", dsgn.mat, grid = FALSE, p = 500,
...)

Arguments

n

Integer. The number of observations.

coord

A vector or matrix that gives the coordinates of each location. Each row corresponds to one location - if any. May be missing if dsgn.mat is specified.

cov.mod

A character string that specifies the max-linear model. Currently only the discretized Smith model is implemented, i.e., cov.mod = "gauss".

dsgn.mat

The design matrix of the max-linear model — see Section Details. May be missing if coord and cov.mod are given.

grid

Logical. Does coord defines a grid?

p

An integer corresponding to the number of unit Frechet random variable used in the max-linear model — see Section Details.

...

The parameters of the max-stable model — see Section Details.

Details

A max-linear process {Y(x)}\{Y(x)\} is defined by

Y(x)=maxj=1,,pfj(x)Zj,xRd,Y(x) = \max_{j=1, \ldots, p} f_j(x) Z_j, \qquad x \in R^d,

where pp is a positive integer, fjf_j are non negative functions and ZjZ_j are independent unit Frechet random variables. Most often, the max-linear process will be generated at locations x1,,xkRdx_1, \ldots, x_k \in R^d and an alternative but equivalent formulation is

Y=AZ,\bf{Y} = A \odot \bf{Z},

where Y={Y(x1),,Y(xk)}\mathbf{Y} = \{Y(x_1), \ldots, Y(x_k)\}, Z=(Z1,,Zp)\mathbf{Z} = (Z_1, \ldots, Z_p), \odot is the max-linear operator, see the first equation, and AA is the design matrix of the max-linear model. The design matrix AA is a k×pk \times p matrix with non negative entries and whose ii-th row is {f1(xi),,fp(xi)}\{f_1(x_i), \ldots, f_p(x_i)\}.

Currently only the discretized Smith model is implemented for which fj(x)=c(p)φ(xuj;Σ)f_j(x) = c(p) \varphi(x - u_j ; \Sigma) where φ(;Σ)\varphi(\cdot; \Sigma) is the zero mean (multivariate) normal density with covariance matrix Σ\Sigma, uju_j is a sequence of deterministic points appropriately chosen and c(p)c(p) is a constant ensuring unit Frechet margins. Hence if this max-linear model is used, users must specify var for one dimensional processes, and cov11, cov12, cov22 for two dimensional processes.

Value

A matrix containing observations from the max-linear model. Each column represents one stations. If grid = TRUE, the function returns an array of dimension nrow(coord) x nrow(coord) x n.

Author(s)

Mathieu Ribatet

References

Wang, Y. and Stoev, S. A. (2011) Conditional Sampling for Max-Stable Random Fields. Advances in Applied Probability.

See Also

condrmaxlin, rmaxstab

Examples

## A one dimensional simulation from a design matrix. This design matrix
## corresponds to a max-moving average process MMA(alpha)
n.site <- 250
x <- seq(-10, 10, length = n.site)

## Build the design matrix
alpha <- 0.8
dsgn.mat <- matrix(0, n.site, n.site)
dsgn.mat[1,1] <- 1

for (i in 2:n.site){
dsgn.mat[i,1:(i-1)] <- alpha * dsgn.mat[i-1,1:(i-1)]
dsgn.mat[i,i] <- 1 - alpha
}

data <- rmaxlin(3, dsgn.mat = dsgn.mat)
matplot(x, t(log(data)), pch = 1, type = "l", lty = 1, ylab =
expression(log(Y(x))))

## One realisation from the discretized Smith model (2d sim)
x <- y <- seq(-10, 10, length = 100)
data <- rmaxlin(1, cbind(x, y), cov11 = 3, cov12 = 1, cov22 = 4, grid =
TRUE, p = 2000)
image(x, y, log(data), col = heat.colors(64))

Simulation of Max-Stable Random Fields

Description

This function generates realisation from a max-stable random field.

Usage

rmaxstab(n, coord, cov.mod = "gauss", grid = FALSE, control =
list(), ...)

Arguments

n

Integer. The number of observations.

coord

A vector or matrix that gives the coordinates of each location. Each row corresponds to one location - if any.

cov.mod

A character string that gives the max-stable model. This must be one of "gauss" for the Smith model, "brown" for the Brown–Resnick model, or "whitmat", "cauchy", "powexp" and "bessel" for the Schlather model with the given correlation family. If the latters are prefix by a "t", e.g., "twhitmat", this would corresponds to the extremal-t model.

grid

Logical. Does the coordinates represent grid points?

control

A named list with arguments 'nlines' (number of lines of the TBM simulation), 'method' the name of the simulation method - must be one of 'exact', 'tbm' or 'circ', and 'uBound' the uniform upper bound. See details.

...

The parameters of the max-stable model. See details.

Details

Users must supply the parameters for the max-stable model. For the Schlather model, users should supply the "nugget", "range" and "smooth" parameter values. For the Smith model, if coord is univariate you must specify var, otherwise users should supply the covariance parameters i.e. parameters with names such as cov11, cov12, ... For the extremal-t model, users should supply the "DoF", "nugget", "range" and "smooth" parameters. Finally for the Brown–Resnick model, users should supply the "range" and the "smooth" parameters.

Here are the details for each allowed components of 'control'. If 'method' is NULL (default), the function tries to find the most appropriate simulation technique. Current simulation techniques are a direct approach, i.e. Cholesky decomposition of the covariance matrix, the turning bands and the circular embedding methods. If 'nlines' is NULL, it is set to 1000. If 'uBound' is NULL, it is set to reasonable values - for example 3.5 for the Schlather model.

Value

A matrix containing observations from the required max-stable model. Each column represents one stations. If grid = TRUE, the function returns an array of dimension nrow(coord) x nrow(coord) x n.

Author(s)

Mathieu Ribatet

References

Schlather, M. (2002) Models for Stationary Max-Stable Random Fields. Extremes 5:1, 33–44.

Smith, R. L. (1990) Max-stable processes and spatial extremes. Unpublished manuscript.

See Also

fitmaxstab

Examples

## 1. Smith's model
set.seed(8)
x <- y <- seq(0, 10, length = 100)
coord <- cbind(x, y)
data <- rmaxstab(1, coord, "gauss", cov11 = 9/8, cov12 = 0, cov22 = 9/8,
  grid = TRUE)
##We change to unit Gumbel margins for visibility
filled.contour(x, y, log(data), color.palette = terrain.colors)

## 2. Schlather's model
data <- rmaxstab(1, coord, cov.mod = "powexp", nugget = 0, range = 3,
smooth = 1, grid = TRUE)
filled.contour(x, y, log(data), color.palette = terrain.colors)

## 3. Brown--Resnick's model **** Only available for non gridded points currently ****
data <- rmaxstab(1, x, cov.mod = "brown", range = 3, smooth = 0.5)
plot(x, log(data), type = "l")

## 4. Extremal-t model *** Very time consuming for 2d grids ***
data <- rmaxstab(1, x, "twhitmat", DoF = 4, nugget = 0, range = 3,
smooth = 0.7)
plot(x, log(data), type = "l")

Analysis of Spatial Extremes

Description

The package SpatialExtremes aims to provide tools for the analysis of spatial extremes. Currently, the package uses the max-stable processes framework for the modelling of spatial extremes.

Max-stable processes are the extension of the extreme value theory to random fields. Consequently, they are good candidate to the analysis of spatial extremes. The strategy used in this package is to fit max-stable processes to data using composite likelihood.

In the future, the package will allow for non-stationarity as well as other approaches to model spatial extremes; namely latent variable and copula based approaches.

A package vignette has been writen to help new users. It can be viewed, from the R console, by invoking vignette("SpatialExtremesGuide").

Details

The package provides the following main tools:

  1. rgp, rmaxstab, rmaxlin, rcopula: simulates gaussian, max-stable, max-linear and copula based random fields,

  2. condrgp, condrmaxlin: conditional simulations for gaussian, max-linear processes,

  3. fitspatgev: fits a spatial GEV model to data,

  4. fitmaxstab, lsmaxstab: fits max-stable processes to data,

  5. latent: draws a Markov chain from a Bayesian hierarchical model for spatial extremes,

  6. predict: allows predictions for fitted max-stable processes,

  7. map, condmap: plot a map for GEV parameter as well as return levels - or conditional return levels

  8. anova, TIC, DIC: help users in model selection,

  9. madogram, fmadogram, lmadogram: are (kind of) variograms devoted to extremes,

  10. fitextcoeff: estimates semi-parametrically the extremal coefficient,

  11. extcoeff: plots the evolution of the extremal coefficient from a fitted max-stable process,

  12. rbpspline: fits a penalized spline with radial basis function,

  13. gev2frech, frech2gev: transform GEV (resp. Frechet) observation to unit Frechet (resp. GEV) ones

  14. gevmle, gpdmle: fit the GEV/GPD distributions to data,

  15. distance: computes the distance between each pair of locations,

  16. profile, profile2d: computes the profile composite likelihood,

  17. covariance, variogram: computes the covariance/semivariogram function.

Acknowledgement

The development of the package has been financially supported by the Competence Center Environment and Sustainability (CCES) and more precisely within the EXTREMES project.

Author(s)

Mathieu Ribatet


Map of the Switzerland.

Description

Plot a map of Switzerland and optionnaly some cities.

Usage

swiss(city = FALSE, add = FALSE, axes = FALSE, km = TRUE, xlab = "",
ylab = "", ...)

Arguments

city

Logical. It TRUE, some city are displayed. Default is to omit.

add

Logical. Should the map be added to an existing plot?

axes

Logical. Should the axis be displayed?

km

Logical. If TRUE (default) the longitude and latitude are expressed in kilometers. Otherwise it is in meters.

xlab, ylab

The x and y-axis labels.

...

Optional arguments to be passed to the plot function.

Value

A graphic window.

Author(s)

Dominik Schaub

Examples

swiss()

Elevation in Switzerland

Description

Data required for plotting a Switzerland map with elevation.

Usage

data(swissalt)

Format

This data set contains three R objects. 'alt.mat' is a 192 by 115 matrix giving the elevation at the grid points defined by 'lon.vec' and 'lat.vec'.

Author(s)

Mathieu Ribatet

Examples

data(swissalt)

layout(matrix(c(1,2), 1), width = c(3.5,1))
mar <- rep(0, 4)
op <- par(mar = mar)
breaks <- seq(0, 2000, by = 250)
image(lon.vec, lat.vec, alt.mat, col = terrain.colors(length(breaks) - 1),
      xaxt = "n", yaxt = "n", bty = "n", xlab = "", ylab = "", xlim =
c(480, 840), ylim = c(58, 300))
swiss(add = TRUE, city = TRUE)

##Heat bar
mar <- c(3, 3, 3, 4)
par(las = 1, mar = mar)

plot.new()
plot.window(xlim = c(0, 1), ylim = range(breaks), xaxs = "i", 
            yaxs = "i")
rect(0, breaks[-length(breaks)], 1, breaks[-1], col = terrain.colors(length(breaks) - 1),
     border = NA)
axis(4, at = breaks[-length(breaks)])
box()
title("Elevation\n(meters)")
par(op)

Detecting spatial trends graphically

Description

This function performs a symbol plot to help in identifying any spatial trends

Usage

symbolplot(data, coord, which = "gev", plot.border = NULL,
 col = c("#FF000080", "#0000FF80"), plot.legend = TRUE, scale = 1)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

which

A character string specifying which values should be plotted. Must be one of "gev" (for the GEV marginal parameters), "mean" for pointwise mean or "median" for pointwise median.

plot.border

An R function that plots the border of the study region. If NULL (default) no border is plotted.

col

A vector of length 2 giving the colors to be used to fill the circles.

plot.legend

Logical. If TRUE (default), a legend is added to the plot.

scale

Positive number. It enables to enlarge (if scale > 1) or reduce (if 0 < scale < 1) the radius of the plotted circles to get a better display.

Details

This function will plot several circles whose center is located at the weather stations and whose radius is proportional to the departure of the value at that position to the areal mean value.

Value

A plot.

Author(s)

Mathieu Ribatet

Examples

## Symbol plot for the Swiss rainfall data set
data(rainfall)
symbolplot(rain, coord, plot.border = swiss)

Takeuchi's information criterion

Description

Computes a the Takeuchi's information criterion which is equivalent to the AIC when the model is miss-specified.

Usage

TIC(object, ..., k = 2)

Arguments

object

An object of class maxstab or spatgev. Often, it will be the output of the fitmaxstab or fitspatgev function.

...

Additional objects of class maxstab or spatgev for which TIC should be computed.

k

Numeric. The penalty per parameter to be used. The case k = 2 (default) correspond to the classical TIC and k=lognk = \log n, n number of observations, is the robust version of the BIC.

Details

TIC is like AIC so that when comparing models one wants to get the lowest TIC score.

Value

Numeric.

Author(s)

Mathieu Ribatet

References

Gao, X. and Song, P. X.-K. (2009) Composite likelihood Bayesian information criteria for model selection in high dimensional data. Preprint.

Sakamoto, Y., Ishiguro, M., and Kitagawa G. (1986) Akaike Information Criterion Statistics. D. Reidel Publishing Company.

Varin, C. and Vidoni, P. (2005) A note on composite likelihood inference and model selection. Biometrika 92(3):519–528.

See Also

fitmaxstab, AIC

Examples

##Define the coordinate of each location
n.site <- 50
locations <- matrix(runif(2*n.site, 0, 100), ncol = 2)
colnames(locations) <- c("lon", "lat")

##Simulate a max-stable process - with unit Frechet margins
data <- rmaxstab(40, locations, cov.mod = "whitmat", nugget = 0.2, range =
30, smooth = 0.5)

M0 <- fitmaxstab(data, locations, "powexp", fit.marge = FALSE)
M1 <- fitmaxstab(data, locations, "cauchy", fit.marge = FALSE)

TIC(M0, M1)
TIC(M0, M1, k = log(nrow(data)))

Summer/Winter annual maxima/minima temperature in continental US.

Description

Summer maxima/Winter minima temperatures over the years 1911–2010 observed at 424 weather stations located in continental USA.

Usage

data(USHCNTemp)

Format

This data set contains three R objects: 'maxima.summer', 'minima.winter' and 'metadata'. 'maxima.summer' is a 100 by 424 matrix giving the temperature in degrees, each column correspond to one location. 'minima.winter is a 99 by 424 matrix giving the temperature in degrees, each column correspond to one location. 'metadata' is a 424 by 5 data frame giving station identifier, the longitude, latitude, elevation and the state for each station.

Author(s)

Mathieu Ribatet

Examples

data(USHCNTemp)
##require(maps) ## <<-- to plot US borders
maps::map("usa")
plot(metadata[,2:3], pch = 15)

Empirical variogram

Description

This function computes the empirical variogram.

Usage

variogram(data, coord, n.bins, xlab, ylab, angles = NULL, add = FALSE,
xlim = c(0, max(dist)), ...)

Arguments

data

A matrix representing the data. Each column corresponds to one location.

coord

A matrix that gives the coordinates of each location. Each row corresponds to one location.

n.bins

The number of bins to be used. If missing, pairwise madogram estimates will be computed.

xlab, ylab

The x-axis and y-axis labels. May be missing. Note that ylab must have the same length has which.

angles

A numeric vector. A partition of the interval (0,π)(0, \pi) to help detecting anisotropy.

add

Logical. If TRUE, the plot is added to the current figure; otherwhise (default) a new plot is computed.

xlim

A numeric vector of length 2 specifying the x coordinate range.

...

Additional options to be passed to the plot function.

Value

A graphic and (invisibly) a matrix with the lag distances and the empirical variogram estimates.

Author(s)

Mathieu Ribatet

See Also

fmadogram, lmadogram

Examples

n.site <- 20
n.obs <- 100
coord <- matrix(runif(2 * n.site, 0, 10), ncol = 2)
data <- rgp(n.obs, coord, "powexp", sill = 2, range = 3, smooth = 1)
variogram(data, coord)

Van der Corput Sequence

Description

This function generates the three dimensional Van der Corput sequence on the half unit Sphere of R3R^3 - eventually randomly rotated.

Usage

vdc(n, rand.rot = FALSE)

Arguments

n

Integer. The length of the sequence.

rand.rot

Logical. Should the sequence be randomly rotated?

Value

A matrix giving the coordinates of the points in the half unit sphere.

Author(s)

Mathieu Ribatet

References

Freulon, X., de Fouquet, C., 1991. Remarques sur la pratique des bandes tournantes a trois dimensions. Cahiers de geostatistique, Fascicule 1, Centre de Geostatistique, Ecole des Mines de Paris, Fontainebleau, pp. 101–117.

Examples

vdc(10)

Annual maxima wind gusts in the Netherlands.

Description

Annual maxima wind gusts (km/h) in the Netherlands recorded at 35 weather stations over the period 1971–2012.

Usage

data(wind)

Format

This data set contains two R objects: 'coord' and 'wind'. 'wind' is a 42 by 35 matrix giving the wind gust speeds in km/h—with missing values, each column correspond to one location. 'coord' is a 35 by 3 matrix giving the longitude (degree), latitude (degree) and elevation (m).

Author(s)

Mathieu Ribatet

Examples

##require(maps)
data(wind)
par(mar = rep(0, 4))
maps::map(xlim = c(0, 9), ylim = c(47.5, 57.5))
points(coord[,1:2], pch = 15)