rcopulaGARCH: Simulation for copula GARCH models with factor structure

In YafeiXu/CopulaModel: CopulaModel: Dependence Modeling with Copulas

Description

Simulation for copula GARCH models with factor structure, with links to C code

Usage

rgarch1fact(n,garchpar,cpar,sigma0,copcode)
rgarch2fact(n,garchpar,cpar,sigma0,copcode)
rgarchbifact(n,grsize,garchpar,cpar,sigma0,copcode)
rgarchnestfact(n,grsize,garchpar,cpar,sigma0,copcode)
rgarchbifactmvt(n,grsize,garchpar,cpar,sigma0,copcode=2) # monotone interpolation
rgarchnestfactmvt(n,grsize,garchpar,cpar,sigma0,copcode=2) # monotone interpolation

Arguments

n	sample size
garchpar	6xd matrix, where d is the number of assets, rows are mu,ar1,omega,alpha1,beta1,nu; set ar1 to be 0 vector if AR term not used in GARCH model
cpar	copula parameter vector
grsize	vector of group sizes for mgrp groups with sum(grsize)=d
sigma0	d-vector with starting values for conditional SDs (from GARCH output)
copcode	current options are 1 for Gaussian, 2 for t, 3 for Gumbel, -3 for reflected Gumbel, 5 for Frank; 9 for BB1 1-factor, 9 for BB1/Frank 2-factor and b-factor; 11 for Gumbel/BB1 nested factor

Details

rgarchbifactmvt and rgarchnestfactmvt use monotone interpolation for the Student t univariate cdf, for faster computations

Value

lgret	dxn matrix of log returns
portfret	nx1 portfolio return vector, assuming equally weighted

See Also

rfactcop structcop

Examples

 
garchpar=matrix(c(0.094, 0.125, 0.068, 0.212, 0.092,
 -0.062,-0.040,-0.082,-0.020,-0.051,
  0.019, 0.014, 0.016, 0.082, 0.014,
  0.071, 0.075, 0.088, 0.092, 0.080,
  0.909, 0.914, 0.882, 0.842, 0.902,
 9.4, 8.3,9.5, 7.9,9.6), 6,5,byrow=TRUE)
sigma0=c(1.113,1.286,0.857,1.176,0.968)
garchpar6=cbind(garchpar,c(0.1, -0.05,0.01,0.08,0.9,10.))
sigma6=c(sigma0,1)
grsize=c(2,2,2)
cpar1=seq(1.1,1.5,.1)
cpar2=c(seq(1.1,1.5,.1),rep(1.1,5))
cparbi=c(seq(1.1,1.6,.1),rep(1.1,6))
cparne=c(rep(1.1,3),seq(1.1,1.6,.1))
#
set.seed(123)
out=rgarch1fact(3,garchpar,cpar1,sigma0,copcode=3)
print(out)
set.seed(123)
out=rgarch2fact(3,garchpar,cpar2,sigma0,copcode=3)
print(out)
set.seed(123)
out=rgarchbifact(3,grsize,garchpar6,cparbi,sigma6,copcode=3)
print(out)
set.seed(123)
out=rgarchnestfact(3,grsize,garchpar6,cparne,sigma6,copcode=3)
print(out)

YafeiXu/CopulaModel documentation built on May 9, 2019, 11:07 p.m.