spec.bt: Blackman-Tukey spectral density estimation

In jrevenaugh/TSAUMN: Time Series Analysis Tools Used in ESCI5201 at the University of Minnesota

Description

Perform Blackman-Tukey power-spectral density estimate of a timeseries.

Usage

spec.bt(s = NA, q = 20, taper = TRUE, n = 2048, deltat = 1,
  demean = TRUE, detrend = FALSE, plot = TRUE, ...)

Arguments

s	univariate real-valued timeseries.
q	maximum lag retained in the auto-correlation function.
taper	apply Akaike ACF taper if true (default).
n	length of zero-padded ACF function.
deltat	sample interval of timeseries if s passed as a vector.
demean	logical. If true, timeseries mean is removed.
detrend	logical. If true, linear trend is removed from timeseries.
plot	logical. If true, spectrum is plotted using gplot.mtm
...	Additional parameters passed to gplot.mtm

Details

Blackman-Tukey is a parametric power-spectral density estimator. It assumes an MA(q) process with known q. In practice, increasing q trades off greater spectral resolution with greater estimator variance. In all cases, q should be much shorter than the timeseries itself.

Value

list with items:

freq	Vector of sampled frequencies
spec	Vector of power spectral estimates
series	Name of the input time series
Method	"Blackman Tukey"

Examples

s <- arimaSim( 1028, ma = 0.5 )
spec <- spec.bt( s, q = 5 )

jrevenaugh/TSAUMN documentation built on Nov. 8, 2019, 2:20 p.m.