biperiodogram {rhosa}

Calculate biperiodogram

Description

Calculate the biperiodogram of real-valued time series

Usage

biperiodogram(
  x,
  dft_given = FALSE,
  mc = FALSE,
  mc_cores = getOption("mc.cores", 2L)
)

Arguments

x

Given time series (or its DFT), as a data frame or matrix with which columns correspond to sampled stretches

dft_given

If TRUE, suppose that DFTs are given instead of time series data and skip the fast fourier transform. Default: FALSE.

mc

If TRUE, calculation is done in parallel computation. Defaults to FALSE.

mc_cores

The number of cores in use for parallel computation, passed parallel::mcmapply() etc. as mc.cores.

Value

A list with names

f1:

The first elements of frequency pairs.

f2:

The second elements of frequency pairs.

value:

The biperiodogram as a matrix. Each of its rows is for a frequency pair; its columns correspond to stretches.

References

Hinich, M.J., 1994. Higher order cumulants and cumulant spectra. Circuits Systems and Signal Process 13, 391–402. doi:10.1007/BF01183737

Examples

f <- function(x) {
    sin(2 * x) + sin(3 * x + 1) + sin(2 * x) * sin(3 * x + 1)
}
v <- sapply(seq_len(1280), f) + rnorm(1280)
m <- matrix(v, nrow = 128)
bp <- biperiodogram(m)

m2 <- stats::mvfft(m)
bp2 <- biperiodogram(m2, dft_given = TRUE)