gammaconvfactor: Gamma convolution factor model

In YafeiXu/CopulaModel: CopulaModel: Dependence Modeling with Copulas

Description

Gamma convolution factor model and its copula

Usage

rgammaconv(n,th0,thvec)
pmgamfact(xvec,th0,thvec,zero=0)
dmgamfact(xvec,th0,thvec,zero=0)
dmgamfact.gl(xvec,th0,thvec,gl)
pbgamfact(x1,x2,th0,th1,th2,zero=0) 
dbgamfact(x1,x2,th0,th1,th2,zero=0)
dbgamfact.gl(x1,x2,th0,th1,th2,gl)
pbgamfcop(u,v,param)
dbgamfcop(u,v,param,zero=0)
dbgamfcop.gl(u,v,param,gl=gldefault)
pcondbgamfcop21(v,u,param,zero=0)
pcondbgamfcop12(u,v,param,zero=0)
pmgamfcop(uvec,param)
dmgamfcop(uvec,param,zero=0)
dmgamfcop.gl(uvec,param,gl)

Arguments

n	sample size for simulation
th0	scalar for shape parameter of the shared/common component
thvec	vector of shape parameters of individual components, length d
xvec	vector of length d with positive values
uvec	vector of length d with values in (0,1)
gl	Gauss-Legendre object with components $nodes and $weights
x1	positive value for first variable (bivariate case)
x2	positive value for second variable (bivariate case)
th1	scalar for shape parameter of first variable (bivariate case)
th2	scalar for shape parameter of second variable (bivariate case)
zero	tolerance for numerical integration, set as 0.0001 if there are problems
u	value in interval 0,1; could be a vector
v	value in interval 0,1; could be a vector
param	parameter vector with length d+1 with th0,thvec

Value

random sample (nxd matrix) for rgammaconv

cdf or conditional cdf or pdf for remaining functions

See Also

invGaussconvfactor

Examples

n=1000
th0=2
thvec=c(.3,.3)
set.seed(123)
xdat=rgammaconv(n,th0,thvec)
cor(xdat)
#plot(xdat)
gl=gausslegendre(35)
pmgamfact(c(1,1.1),th0,c(.4,.4),zero=0)
dbgamfact(1,1.1,th0,th1=.4,th2=.4,zero=0)
dbgamfact.gl(1,1.1,th0,th1=.4,th2=.4,gl) # could be inaccurate for th1,th2<1
dbgamfact(1,1.1,th0,th1=1.2,th2=1.4,zero=0)
dbgamfact.gl(1,1.1,th0,th1=1.2,th2=1.4,gl) 
dmgamfact(c(1,1.1),th0,c(1.2,1.4),zero=0)
dmgamfact.gl(c(1,1.1),th0,c(1.2,1.4),gl)
# density can be finite on diagonal
try(dbgamfact(1,1.0001,0.1,th1=.4,th2=.4,zero=0))
try(dbgamfact(1,1,0.1,th1=.4,th2=.4,zero=0))
# copula 
pmgamfcop(c(.5,.6),c(2,1.2,1.4))
dmgamfcop(c(.5,.6),c(2,1.2,1.4),zero=0)
dmgamfcop.gl(c(.5,.6),c(2,1.2,1.4),gl)
pcondbgamfcop21(.6,.5,c(2,1.2,1.4),zero=0)
pcondbgamfcop12(.5,.6,c(2,1.2,1.4),zero=0)
pcondbgamfcop21(.5,.6,c(2,1.4,1.2),zero=0)

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