ewspecHaarNonPer: Compute evolutionary wavelet spectrum (EWS) estimate based on the Haar wavelet transform.

Description

This function uses the special HwdS function to compute the Haar wavelet transform with out boundary conditions (neither periodic, interval, mirror reflection). This is so all coefficients are genuine Haar coefficients without involving extra/repeated data.

Usage

ewspecHaarNonPer(x, filter.number = 1, family = "DaubExPhase",
    UseLocalSpec = TRUE, DoSWT = TRUE, WPsmooth = TRUE,
    verbose = FALSE, smooth.filter.number = 10,
    smooth.family = "DaubLeAsymm",
    smooth.levels = 3:WPwst$nlevels - 1, smooth.dev = madmad,
    smooth.policy = "LSuniversal", smooth.value = 0,
    smooth.by.level = FALSE, smooth.type = "soft",
    smooth.verbose = FALSE, smooth.cvtol = 0.01,
    smooth.cvnorm = l2norm, smooth.transform = I,
    smooth.inverse = I)

Value

The same value as for the ewspec function.

Arguments

x: A vector of dyadic length that contains the time series you want to form the EWS of.
filter.number: Should always be 1 (for Haar)
family: Should always be "DaubExPhase", for Haar.
UseLocalSpec: Should always be TRUE.
DoSWT: Should always be TRUE
WPsmooth: Should alway be TRUE to do smoothing. If FALSE then not smoothed.
verbose: If TRUE informative messages are printed during the progress of the algorithm.
smooth.filter.number: Wavelet filter number for doing the wavelet smoothing of the EWS estimate.
smooth.family: Wavelet family for doing the wavelet smoothing of the EWS estimate.
smooth.levels: Which levels of the EWS estimate to apply smoothing to.
smooth.dev: What kind of deviance to use. The default is madmad, an alternative might be var.
smooth.policy: What kind of smoothing to use. See help page for ewspec
smooth.value: If a manual value has to be supplied according to the smooth.policy then this is it.
smooth.by.level: If TRUE then all levels are smoothed independently with different smoothing, otherwise all levels are smoothed together (eg one threshold for all levels).
smooth.type: The type of wavelet smoothing "hard" or "soft"
smooth.verbose: If TRUE then informative messages about the smoothing are printed.
smooth.cvtol: If cross-validation smoothing is used, this is the tolerance
smooth.cvnorm: If cross-validation smoothing used, this is the norm that's used
smooth.transform: A transform is applied before smoothing
smooth.inverse: The inverse transform is applied after smoothing

Author

Guy Nason.

Details

This function is very similar to ewspec from wavethresh, and many arguments here perform the same function as there.

References

Nason, G.P. (2013) A test for second-order stationarity and approximate confidence intervals for localized autocovariances for locally stationary time series. J. R. Statist. Soc. B, 75, 879-904. tools:::Rd_expr_doi("10.1111/rssb.12015")

Examples

Run this code

#
# Requires wavethresh, so not run directly in installation of package
#
ewspecHaarNonPer(rnorm(512))

Run the code above in your browser using DataLab