NatafInt: Solve the Nataf integral with an alternative integration...

In itsoukal/anySim: Stochastic Simulation of Processes with any Marginal Distribution and Correlation Structure

Description

Estimation of the resulting (i.e., in the actual domain) correlation coefficients, given the equivalent correlation coefficients (i.e., in the Gaussian domain).

Usage

NatafInt(rho, fx, fy, paramlistfx, paramlistfy)

Arguments

rho	A scalar or vector of correlation coefficients (i.e., in seq(from=0, to=1, by=0.1)).
fx	A string indicating the quantile function of the distribution (i.e., the ICDF).
fy	A string indicating the quantile function of the distribution (i.e., the ICDF).
paramlistfx	A named list with the parameters of the distribution.
paramlistfy	A named list with parameters of the distribution.

Value

A vector of correlation coefficients in the actual domain.

Examples

## The case of two identrical Gamma distributions, with shape=1 and scale=1.
## Not run: 
fx=fy='qgamma'
pfx=pfy=list(shape=1, scale=1)
rhoz=seq(from=0, to=1 , by=0.2)
rhox=NatafInt(rho = rhoz, fx = fx, fy = fy, paramlistfx = pfx, paramlistfy = pfy)
plot(rhoz,rhox); abline(0,1)

## The case identrical Bernoulli distributions, with size=1 and prob=0.3.
fx=fy='qbinom'
pfx=pfy=list(size=1, prob=0.3)
rhoz=seq(from=0, to=1 , by=0.2)
rhox=NatafInt(rho = rhoz, fx = fx, fy = fy, paramlistfx = pfx, paramlistfy = pfy)
plot(rhoz,rhox); abline(0,1)

## The case of two identrical zero-inflated (i.e., mixed) distributions,
## with p0=0.7 a Gamma distribution
## for the continuous part with shape=1 and scale=1.

fx=fy='qzi'
pfx=pfy=list(Distr=qgamma, p0=0.7, shape=1, scale=1)
rhoz=seq(from=0, to=1 , by=0.2)
rhox=NatafInt(rho = rhoz, fx = fx, fy = fy, paramlistfx = pfx, paramlistfy = pfy)
plot(rhoz,rhox);abline(0,1)
# compare with the Gauss-Hermite integration method
rhox=NatafGH(rho = rhoz, fx = fx, fy = fy, paramlistfx = pfx, paramlistfy = pfy, nodes = 21)
points(rhoz,rhox,col='red',pch=19)

## End(Not run)

itsoukal/anySim documentation built on May 7, 2020, 11:57 p.m.