astsa (version 2.4)

SigExtract: Signal Extraction And Optimal Filtering

Description

Performs signal extraction and optimal filtering as discussed in Chapter 4.

Usage

SigExtract(series, L = c(3, 3), M = 50, max.freq = 0.05, col = 4)

Value

Returned invisibly as a list

series.filt

[[1]] the filtered series

filter

[[2]] the filter

Also prints the rounded filter coefficients and returns plots of (1) the original and filtered series, (2) the estimated spectra of each series, (3) the filter coefficients and the desired and attained frequency response function.

Arguments

series

univariate time series to be filtered

L

degree of smoothing (may be a vector); see spans in spec.pgram for more details

M

number of terms used in the lagged regression approximation

max.freq

truncation frequency, which must be larger than 1/M

col

color of the main graphs

Author

D.S. Stoffer

Details

The basic function of the script, and the default setting, is to remove frequencies above 1/20 (and, in particular, the seasonal frequency of 1 cycle every 12 time points). The sampling frequency of the time series is set to unity prior to the analysis.

References

You can find demonstrations of astsa capabilities at FUN WITH ASTSA.

The most recent version of the package can be found at https://github.com/nickpoison/astsa/.

In addition, the News and ChangeLog files are at https://github.com/nickpoison/astsa/blob/master/NEWS.md.

The webpages for the texts and some help on using R for time series analysis can be found at https://nickpoison.github.io/.

Examples

if (FALSE) {

SigExtract(detrend(soi)) -> u
tsplot(cbind(soi.filtered=u[[1]], detrend(soi)), col=4*1:2, lwd=2:1, spag=TRUE, addLegend=TRUE)

}