fitSpCopula: Spatial Copula Fitting

In spcopula: Copula Driven Analysis - Multivariate, Spatial, Spatio-Temporal

Description

A bivariate spatial copula is composed out of a set of bivariate copulas. These are combined using a convex linear combination with weights based on distances where for copulas with a 1-1 correspondence of Kendall's tau or Spearman's rho a dependence function providing measures of association based on distances might be used. This function estimates a spatial dependence function, evaluates the log-likelihood per family and lag class, selects the best fits and composes a spatial bivariate copula.

Usage

fitSpCopula(bins, data, cutoff = NA, families = c(normalCopula(), tCopula(),
            claytonCopula(), frankCopula(), gumbelCopula()), ...)

Arguments

bins	the bins to be used, typically output from calcBins.
data	the spatial dataset that ahs been used to generate bins.
cutoff	The maximal distance to be used in the fit.
families	The set of families to be investigated.
...	Passed on to the function fitCorFun.

Value

A spCopula object.

Author(s)

Benedikt Graeler

References

Graeler, B. & E. Pebesma (2011): The pair-copula construction for spatial data: a new approach to model spatial dependency. Poster at: Spatial Statistics 2011 - Mapping global change. Enschede, The Netherlands, 23-25 March 2011. DOI: 10.1016/j.proenv.2011.07.036

See Also

Take a look at fitCorFun, loglikByCopulasLags, composeSpCopula and spCopula to gain more control over the fitting procedure.

Examples

# reload some spatial data
library("sp")
data("meuse")
coordinates(meuse) <- ~x+y

# drop margins
meuse$marZinc <- plnorm(meuse$zinc, mean(log(meuse$zinc)), sd(log(meuse$zinc)))

# load data from a provided binning
data("spCopDemo")

fitSpCopula(bins, meuse, 600)

spcopula documentation built on May 2, 2019, 4:49 p.m.