Model checking of a fitted copula based model.

Share:

Description

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

Usage

1
2
## 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

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
## 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)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.