Covariance matrix of \(\hat{\eta}\) for fractional Gaussian noise (fGn).
CetaFGN(eta, m = 10000, delta = 1e-9)
parameter vector eta = c(H, *).
eta = c(H, *)
integer specifying the length of the Riemann sum, with step size 2 * pi/m. The default (10000) is realistic.
2 * pi/m
step size for numerical derivative computation.
Variance-covariance matrix of the estimated parameter vector \(\hat{\eta}\).
Currently, the step size for numerical derivative is the same in all coordinate directions of eta. In principle, this can be far from optimal.
eta
specFGN
# NOT RUN { (C.7 <- CetaFGN(0.7, m = 256)) (C.5 <- CetaFGN(eta = c(H = 0.5), m = 256)) (C.5. <- CetaFGN(eta = c(H = 0.5), m = 1024)) # }
Run the code above in your browser using DataCamp Workspace