surrog {wsyn}

Creates surrogate time series, either Fourier surrogates or amplitude adjusted Fourier surrogates

Description

For significance testing wavelet coherence and other purposes

Usage

surrog(dat, nsurrogs, surrtype, syncpres)

Arguments

dat

A locations x time matrix of observations (for multiple-time series input), or a single vector

nsurrogs

The number of surrogates to produce

surrtype

Either "fft" (for Fourier surrogates) or "aaft" (for amplitude adjusted Fourier surrogates). Fourier surrogates are appropriate for time series with normal marginals; otherwise consider aaft surrogates.

syncpres

Logical. TRUE for "synchrony preserving" surrogates (same phase randomizations used for all time series). FALSE leads to independent phase randomizations for all time series.

Details

Fourier surrogates are somewhat faster than aaft surrogates, and may be much faster when some of the time series in the data have ties. Prenormalization (e.g., using cleandat) can make it possible to use fft surrogates.

Value

surrog returns a list of nsurrogs surrogate datasets

Author(s)

Jonathan Walter, [e-mail]; Lawrence Sheppard, [e-mail]; Daniel Reuman, [e-mail]

References

Sheppard, LW, et al. (2016) Changes in large-scale climate alter spatial synchrony of aphid pests. Nature Climate Change. DOI: 10.1038/nclimate2881

Schreiber, T and Schmitz, A (2000) Surrogate time series. Physica D 142, 346-382.

Prichard, D and Theiler, J (1994) Generating surrogate data for time series with several simultaneously measured variables. Physical Review Letters 73, 951-954.

See Also

wpmf, coh, wlmtest, synmat, browseVignettes("wsyn")

Examples

times<-1:100
dat<-sin(2*pi*times/10)
nsurrogs<-10
surrtype<-"fft"
syncpres<-TRUE
res<-surrog(dat,nsurrogs,surrtype,syncpres)