plotPseudospectrum: Plot and calculate the pseudospectrum of spectrally...

In spectral.methods: Singular Spectrum Analysis (SSA) Tools for Time Series Analysis

Description

This function plots the pseudospectrum of the results from a SSA run, e.g. it plots the variance of the individual eigentriples vs. their frequencies. It can also be used to compute the frequency, variance and period of all SSA eigentriples.

Usage

plotPseudospectrum(ssa.object, calc.raw.SSA = TRUE, plot.spectrum = TRUE, 
    plot.fourier = TRUE, series.orig = c(), pch = 16, col = "red", 
    show.harmonies = TRUE, label.points = TRUE, label.tresh = 5e-04, 
    call.freq = quote(DetermineFreq(series.t)), print.stat = TRUE, 
    ...)

Arguments

ssa.object	SSA object: the results of a run of ssa().
calc.raw.SSA	logical: Whether to additionally compute the whole spectrum for all un-grouped eigentriples (my slow the process in case of long time series.
plot.spectrum	logical: whether to plot the pseudospectrum.
plot.fourier	logical: Whether to plot the Fourier spectrum in the background.
series.orig	numeric vector: original, non decomposed time series (used to calculate Fourier spectrum). If not supplied, an object with the saved in ssa.object is searched for in all active environments.
pch	integer: graphic parameter passed to plot() (?par)
col	character: graphic parameter passed to plot() (?par)
show.harmonies	logical: whether to mark the positions of the harmonies of the oscillation with the highest variance.
label.points	logical: whether to label the points with period values
label.tresh	numeric: threshold used to label points
call.freq	function to use to determine the frequencies of the ssa eigentriples.
print.stat	logical: whether to print status information during the calculations.
...

Value

list with values

paired	matrix: frequency, period and variance of the paired reconstruction
raw	matrix: frequency, period and variance of the unpaired reconstruction

Author(s)

Jannis v. Buttlar

See Also

ssa

spectral.methods documentation built on May 1, 2019, 6:36 p.m.