dduCopula: partial derivatives of copulas

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

Description

Similar to the dCopula and pCopula the function dduCopula evaluates the partial derivative \frac{\partial}{\partial u} C(u,v) and the function ddvCopula evaluates the partial derivative \frac{\partial}{\partial v} C(u,v) of the provided copula.

Usage

dduCopula(u, copula, ...)
ddvCopula(u, copula, ...)

Arguments

u	Pairs of values for which the partial derivative should be evaluated.
copula	The copula object representing the family member of interest.
...	additional arguments can be passed on to the underlying functions.

Value

A vector of the evaluated partial derivatives of the same length as rows in u.

Author(s)

Benedikt Graeler

Examples

####################################
## Asymmetric vs. Gaussian copula ##
####################################

asCop <- asCopula(c(-2,1))
asCopSmpl <- rCopula(100,asCop)

unitScatter(smpl=asCopSmpl)

# conditional probabilities of an asymmetric copula given u
asGivenU <- dduCopula(asCopSmpl,asCop)

# vs. conditional probabilities of an asymmetric copula given v
asGivenV <- ddvCopula(asCopSmpl[,c(2,1)],asCop)
unitScatter(smpl=cbind(asGivenU, asGivenV))

normalCop <- normalCopula(.6)
normCopSmpl <- rCopula(100,normalCop)

unitScatter(smpl=normCopSmpl)

# conditional probabilities of a Gaussian copula given u
normGivenU <- dduCopula(normCopSmpl,normalCop)

# vs. conditional probabilities of a Gaussian copula given v
normGivenV <- ddvCopula(normCopSmpl[,c(2,1)],normalCop)
unitScatter(smpl=cbind(normGivenU, normGivenV))

BenGraeler/spcopula documentation built on Nov. 20, 2020, 4:07 p.m.