uni.vola.sim {ccgarch} | R Documentation |
This function simulates an univariate time series with a GARCH(1,1) volatility.
uni.vola.sim(a, nobs, d.f=Inf, cut=1000)
a |
a vector of parameters (3 times 1) |
nobs |
a number of observations simulated (T) |
d.f |
degrees of freedom parameter for t-distribution |
cut |
a number of observations to be removed to minimise the initial effects |
A list with components:
h |
a GARCH(1,1) volatility (T times 1) |
eps |
a series of error term with the conditional variances "h" (T times 1) |
When "d.f=Inf", the innovations (the standardised residuals) follow the standard normal distribution. Otherwise, they follow a student's t-distribution with the degrees of freedom equal to "d.f".
Bollerslev, T. (1986), “Generalized Autoregressive Conditional Heteroskedasticity”, Journal of Econometrics, 31, 307–327.
Fiorentini, G., G. Calzolari and L. Panattoni (1996), “Analytic Derivatives and the Computation of GARCH Estimates”, Journal of Applied Econometrics, 11, 399–417.
nobs <- 1000 nu = 8 a = c(0.1,0.2,0.7) # a <- c(a constant, ARCH parameter, GARCH parameter) # with normal innovations eps = uni.vola.sim(a, nobs) # with t innovations eps.t = uni.vola.sim(a, nobs, d.f = df)