FGN (version 2.0-12)

FitFGN: MLE estimation for FGN

Description

Exact MLE estimation for FGN

Usage

FitFGN(z, demean = TRUE, MeanMLEQ = FALSE, lag.max = "default")

Arguments

z

time series, vector or ts object.

demean

if True, subtract mean. Otherwise assume it is zero.

MeanMLEQ

if True, an iterative algorithm is used for exact simultaneous MLE estimation of the mean and other parameters.

lag.max

the residual autocorrelations are tabulated for lags 1, ..., lag.max. Also lag.max is used for the Ljung-Box portmanteau test.

Value

A list with class name "FitAR" and components:

loglikelihood

value of the loglikelihood

H

estimate of H parameter

SEH

SE of H estimate

sigsqHat

innovation variance estimate

muHat

estimate of the mean

SEmu

SE of mean

Rsq

R-squared, coefficient of forecastability

LjungBox

table of Ljung-Box portmanteau test statistics

res

normalized residuals, same length as z

demean

TRUE if mean estimated otherwise assumed zero

IterationCount

number of iterations in mean mle estimation

MLEMeanQ

TRUE if mle for mean algorithm used

tsp

tsp(z)

call

result from match.call() showing how the function was called

DataTitle

returns attr(z,"title")

Details

The exact loglikelihood function is maximized numerically using optimize. The standard error for the H parameter is estimated (McLeod, Yu and Krougly, 2007).

References

McLeod, A.I., Yu, Hao, Krougly, Zinovi L. (2007). Algorithms for Linear Time Series Analysis, Journal of Statistical Software.

See Also

GetFitFGN, FitRegressionFGN, Boot.FitFGN, coef.FitFGN, plot.FitFGN, print.FitFGN, summary.FitFGN, HurstK

Examples

data(NileMin)
out<-FitFGN(NileMin)
summary(out)
plot(out)
coef(out)