earfima: Exact MLE for ARFIMA

In FGN: Fractional Gaussian Noise and power law decay time series model fitting

Description

The time series is corrected for the sample mean and then exact MLE is used for the other parameters.

Usage

earfima(z, order = c(0, 0, 0), lmodel = c("FD", "FGN", "PLA", "NONE"))

Arguments

z	time series
order	(p,d,q) where p=order AR, d=regular difference, q=order MA
lmodel	type of long-memory component: FD, FGN, PLA or NONE

Details

The sample mean is asymptotically efficient.

Value

list with components:

bHat	transformed optimal parameters
alphaHat	estimate of alpha
HHat	estimate of H
dHat	estimate of d
phiHat	estimate of phi
thetaHat	estimate of theta
wLL	optimized value of Whittle approximate log-likelihood
LL	corresponding exact log-likelihood
convergence	convergence indicator
algorithm	optimization algorithm used, 1 for L-BFGS-B, 2 for Nelder-Mead, 3 for SANN

Author(s)

Justin Veenstra and A. I. McLeod

Examples

z <- rnorm(100)
earfima(z, lmodel="FGN")

FGN documentation built on May 30, 2017, 7:19 a.m.