FGN (version 2.0-12)

SimulateFGN: Simulates FGN

Description

A fractional Gaussian noise time series is simulated.

Usage

SimulateFGN(n, H)

Arguments

n

length of time series

H

Hurst coefficient

Value

vector of length containing the simulated time series

Details

The FFT is used so it is most efficient if you select n to be a power of 2.

References

Davies, R. B. and Harte, D. S. (1987). Tests for Hurst Effect. Biometrika 74, 95--101.

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

See Also

DLSimulate

Examples

#Example 1
#simulate a process with H=0.2 and plot it
z<-SimulateFGN(100, 0.2)
ts.plot(z)
# 
#Example 2
#simulate FGN and compare theoretical and sample autocovariances
H<-0.7
n<-8192
z<-SimulateFGN(n, H)
#autocovariances
sacvf<-acf(z, plot=FALSE,type="covariance")$acf
tacf<-acvfFGN(H, n-1)
tb<-matrix(c(tacf[1:10],sacvf[1:10]),ncol=2)
dimnames(tb)<-list(0:9, c("Tacvf","Sacvf"))
tb