H: Copula H and G functions

In samorso/nfcm: Spline Based Nonlinear Factor Copula Model

Copula H and G functions

Description

Copula H and G functions

Usage

H(u, v, copula = "normal", param = list())

G(u, w, copula = "normal", param = list())

Arguments

u	latent uniform(0,1) variable.
v	uniform(0,1) errors.
copula	model (see 'Details').
param	list of parameters to pass to the model.
w	uniform(0,1) observable variables

Details

Currently we support the following model: "normal", "student", "skewt" (skew-t distribution), "ghyp" (generalized hyperbolic distribution), "clayton", "gumbel". The list of param depends on the model:

• normal: either correlation corr (between -1 and 1) or variances of Z var_z>0 and \epsilon var_e>0;

• student: degrees of freedom nu>0;

• skewt: degrees of freedom nu>0, the variance of \epsilon var_e>0 and the parameter of asymetry lambda (betweem -1 and 1);

• ghyp: lambda, chi (positive), psi (positive), the variance of \epsilon var_e>0 and gamma;

• clayton: alpha (positive), the parameter of dependence;

• gumbel: alpha>1, the parameter of dependence.

Value

A vector or matrix of the same dimension as v for H or as w for G.

samorso/nfcm documentation built on Oct. 13, 2024, 9:35 p.m.