densi.curv.cop: Computes the density of level curves for copulae models

In mobirep: Models Bivariate Dependence and Produces Bivariate Return Periods

Description

Computes the density along the level curve estimate with a copula. Based on the density function of the selected copula.

Usage

densi.curv.cop(lines, copi, pxf, pyf, u)

Arguments

lines	location of the objective level curve for which the density needs to be estimated in the 2D space
copi	a copula function with the parameters fitted to the bivariate dataset
pxf	uniform values of the 1st margin
pyf	uniform values of the 1st margin
u	original data

Value

density for each points (couple x,y) along the level curves for copulae

See Also

dcopula

Examples

data(porto)

tr1=0.9
tr2=0.9
fire01meantemp=na.omit(fire01meantemp)
u=fire01meantemp

#Compute uniform margins
marg=Margins.mod(tr1,tr2,u=fire01meantemp)
kk=marg$uvar
pp=marg$uvar_ext
uu=marg$val_ext

#Copula parameters
c1=1.5
copu<-copBasic::GHcop
upobj=0.001
interh="comb"

#compute the curve on 3 subdomains
cl1<-curve.funct(pxf=pp[,1],pyf=pp[,2],mar1=uu[,1],mar2=uu[,2],pos="l",
pobje=upobj,ng=100,inter=interh,coco=copu,c1=c1)
cl2<-curve.funct(pxf=pp[,1],pyf=pp[,2],mar1=uu[,1],mar2=uu[,2],pos="m",
pobje=upobj,ng=100,inter=interh,coco=copu,c1=c1)
cl3<-curve.funct(pxf=pp[,1],pyf=pp[,2],mar1=uu[,1],mar2=uu[,2],pos="r",
pobje=upobj,ng=100,inter=interh,coco=copu,c1=c1)

cl<-rbind(cl1,cl2,cl3)

# Homogenization of the number of points
cli<-digit.curves.p(start=c(cl[1,1],cl[1,2]), as.matrix(cl), nPoints=98, closed = FALSE)

# Computes the density along the curve
co=copula::gumbelCopula(c1,dim=2)
cli<-densi.curv.cop(lines=cli,copi=co,pxf=kk[,1],pyf=kk[,2],u=u)

mobirep documentation built on April 22, 2021, 5:07 p.m.