SimulateGaussianAR(FitAR)R Documentation

~~function to do ... ~~

Description

Simulate a mean-zero stationary Gaussian AR(p) time series.

Usage

SimulateGaussianAR(phi, n = 100, InnovationVariance = 1)

Arguments

phi vector containing AR coefficients
n length of time series
InnovationVariance innovation variance

Details

The p initial values are simulated using the appropriate multivariate distribution as was suggested in McLeod (1975). The R function rnorm is used.

Value

a vector of length n, the simulated series

Note

If the process is non-stationary, then random initial values are used determined by the first p innovations.

Author(s)

A.I. McLeod

References

McLeod, A.I. (1975), Derivation of the theoretical autocorrelation function of autoregressive moving-average timeseries, Applied Statistics 24, 255-256. Percival, D.B. and Walden, A.T. (1993). Spectral Anaysis for Physical Applications.

See Also

boot.FitAR

Examples

#Percival and Walden (1993, p.46) illustrated a time series with a 
#very peaked spectrum with the AR(4) with coefficients
#c(2.7607,-3.8106,2.6535,-0.9238) with NID(0,1) innovations.
#
z<-SimulateGaussianAR(c(2.7607,-3.8106,2.6535,-0.9238),1000)
library(lattice)
TimeSeriesPlot(z)

[Package FitAR version 1.0 Index]