Simulate linear time series

Share:

Description

The Durbin-Levinsion recursions are used to simulate a stationary time series given an unit innovation sequence and given autocovariance function. Requires

O(n^2)

flops.

Usage

1
DLSimulate(n, r, useC = TRUE, rand.gen = rnorm, ...)

Arguments

n

length of time series to be generated

r

autocovariances, lags 0, ...,

useC

=TRUE, use C interface. Otherwise direct computation.

rand.gen

random number generator to use

...

optional arguments passed to rand.gen

Details

See Hipel and McLeod (1994) or McLeod, Yu and Krougly (2007).

Value

simulated time series of length n

Author(s)

A.I. McLeod

References

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

See Also

DHSimulate, SimGLP, codearima.sim

Examples

1
2
3
4
5
6
7
8
9
#Simulate hyperbolic decay time series 
#with Hurst coefficient, H=0.9
n<-2000
H<-0.9
alpha<-2*(1-H)  #hyperbolic decay parameter
r<-(1/(1:n))^alpha
z<-DLSimulate(n, r)
plot.ts(z)
#can use HurstK function in FGN library to estimate H