Scalar BEKK(1,1) multivariate GARCH — morie_bekk

\(H_t = C'C + a \epsilon_{t-1}\epsilon_{t-1}' + b H_{t-1}\) with variance targeting, so \(H_t\) is positive definite by construction. Mirrors morie.fn.mgrch.

Usage

morie_bekk_garch(R)

Arguments

R: matrix (T x k), k >= 2.

Value

list: H (T x k x k array), a, b, persistence, H_bar, loglik, T, k.

References

Engle, R. F. & Kroner, K. F. (1995). Multivariate simultaneous generalized ARCH. Econometric Theory, 11(1), 122-150.

Examples

set.seed(1)
morie_bekk_garch(matrix(rnorm(200), ncol = 2))$persistence
#> [1] 0.9986521