mtvgarchSim: Simulate from a multivariate TV-GARCH-X model

Description

Simulate from a multivariate multiplicative TV(s)-GARCH(p,q,r)-X model.

Usage

mtvgarchSim(n, m = 2, order.g = c(1,1), order.h = c(1,1,0, 1,1,0), 
order.x = NULL, intercept.g = c(1.2,1), size = c(3,5), speed = c(10,25), 
location = c(0.5,0.8), intercept.h = c(0.2,0.3), arch = c(0.10,0.05), 
garch = c(0.80,0.90), asym = NULL, xtv = NULL, xreg = NULL, par.xreg = NULL,  
R = c(1,0.6,0.6,1), dcc = FALSE, par.dcc = NULL, opt = 0, as.zoo = TRUE, 
verbose = FALSE, innovations = NULL)

Value

An object of class 'zoo' (if as.zoo = TRUE), otherwise a matrix or a list (if verbose = TRUE), with the simulated values.

Arguments

n: integer.
m: integer indicating the dimenson of the multivariate series.
order.g: integer matrix with each row indicating the number of locations in each transition function of the TV components; m rows and max.s columns.
order.h: integer matrix with each row indicating the order.h for each series; the first column controls the GARCH order, the second the ARCH order and the third the asymmetry order of the GARCH-X components.
order.x: NULL or binary matrix indicating which xreg variables should be included as covariates in the GARCH-X components. If provided and xreg is NULL, volatility spillovers for the selected series are included as covariates.
intercept.g: NULL or numeric vector.
size: NULL or numeric matrix containing the size coefficients. Only relevant for TV-GARCH models.
speed: NULL or numeric matrix containing the speed coefficients. Only relevant for TV-GARCH models.
location: NULL or numeric matrix containing the location coefficients; m rows and max.c columns. Only relevant for TV-GARCH models.
intercept.h: numeric matrix.
arch: NULL or numeric matrix containing the ARCH coefficients.
garch: NULL or numeric matrix containing the GARCH-type coefficients.
asym: NULL or numeric matrix containing the asymmetry-type coefficients.
xtv: NULL or numeric vector, time series or zoo object to include as the transition variable in the TV component. If NULL, calendar time, scaled between 0 and 1, is used as the transition variable. Only relevant for TV-GARCH models.
xreg: numeric vector, matrix, time series or zoo object to include as covariates in the GARCH-X component.
par.xreg: NULL or numeric matrix containing the covariates initial coefficients.
R: matrix of (constant) conditional correlations.
dcc: logical. If TRUE, dynamic conditional correlations are estimated. If FALSE (default), then the conditional correlations are constant.
par.dcc: numeric vector containing the ARCH- and GARCH-type coefficients in the dynamic conditional correlations.
opt: integer indicating whether the speed parameter in the TV component should be scaled. If 0, no scaling; if 1, speed/sd(xtv); if 2, exp(speed). Only relevant for TV-GARCH models.
as.zoo: logical. If TRUE, then the returned result is of class zoo.
verbose: logical, if TRUE, the conditional variance and the innovations are also returned.
innovations: NULL or numeric matrix with the innovations. If NULL, then standard normal innovations are generated with rnorm.

Author

Susana Campos-Martins

Examples

Run this code

set.seed(12345)

## Simulate from a bivariate CCC-TV(1)-GARCH(1,1) model (default):
mySim1 <- mtvgarchSim(n = 1500)

## Simulate from a bivariate CCC-TV(1)-GARCH(1,1)-X model
## (with volatility spillovers)
mySim2 <- mtvgarchSim(n = 1500, order.x = c(0,1,1,0), par.xreg = 
c(0.03,0.04))