ebeedcc: Estimation of DCC models, equation by equation

In TaoussD/EbEEMGARCH: Estimating MGARCH(1,1) model equation by equation

Description

3 steps estimation of DCC models equation by equation

Usage

estimDCC.EbEE(Omega, A, B, S, alpha, beta, eps, r = 10,type)

Arguments

Omega	Initialization paramater for Omega (vector)
A	Initialization parameter for A (matrix)
B	Initialization parameter for B (vector)
S	Initialization parameter for S, correlation matrix
alpha	Initialization parameter for alpha, first DCC parameter
beta	Initialization parameter for beta, second DCC parameter
eps	m*n matrix of the data
r	Number of observations used as initialisation parameter
type	type="Engle" for estimation as an Engle-DCC type="Aielli" for estimation as an Aielli-DCC

Value

Omega	Estimation parameter for Omega
A	Estimation parameter for A (first step)
B	Estimation parameter for B (first step)
alpha	Estimation parameter for alpha (second step)
beta	Estimation parameter for beta (second step)
S	Estimation parameter for the correlation matrix (third step)

Author(s)

D. Taouss & C. Francq

References

C. Francq & J.M. Zakoian, Estimating multivariate GARCH and Stochastic Correlation models equation by equation, October 2014
G.P. Aielli, Dynamic Conditional Correlation: on Properties and Estimation, July 2011

See Also

EbEEMGARCH Homepage of the documentation

Examples

#Sampling some data
m<-2
Omega <- c(1, 1);
A <- matrix(rep(0.025, m ^ 2), ncol = m)
B <- c(0.8, 0.8);
S <- matrix(c(1, 0.3, 0.3, 1), nrow = 2)
alpha <- 0.05;
beta <- 0.99 - alpha
n <- 2500
nu <-7

eps <- GarchDCC.sim(n, Omega, A, B, alpha, beta, S, nu = nu, model="Aielli",noise = "student")

#Estimating parameters
Omegainit <- rep(0.02, m)
Ainit <- matrix(rep(0.03, m ^ 2), ncol = m)
Binit <- rep(0.7, m)
Sinit<-diag(rep(1,m))
alphainit <- 0.04
betainit <- 0.97 - alphainit

res <- estimDCC.EbEE(Omegainit, Ainit, Binit, Sinit, alphainit, betainit, eps$sim, type="Aielli")

TaoussD/EbEEMGARCH documentation built on May 9, 2019, 4:18 p.m.