HaarConcat: Generate a concatenated Haar MA process

In wavethresh: Wavelets Statistics and Transforms

Generate a concatenated Haar MA process

Description

This function generates a particular set of four concatenated Haar MA processes.

Usage

HaarConcat()

Arguments

None

Details

This function generates a realization of particular kind of non-stationary time series probability model. The returned time series is the result of concatenating 4 time series each of length 128 from the Haar MA process generator (HaarMA) of orders 1, 2, 3 and 4. The standard deviation of the innovations is 1. This function was used to generate the figure of the concatenated Haar MA process in Nason, von Sachs and Kroisandt. It produces a kind of time series that can be sparsely represented by the wavelet machinery but at the same time is non-stationary.

See Nason, von Sachs and Kroisandt (2000) Wavelet processes and adaptive estimation of the evolutionary wavelet spectrum. J R Statist Soc, B, 62, 271-292.

Value

A vector containing 512 observations from four concatenated Haar MA processes

Author(s)

G P Nason

See Also

HaarMA,ewspec

Examples

#
# Generate the concatenated Haar MA process.
#
MyHaarCC <- HaarConcat()
#
# Plot it
#
## Not run: ts.plot(MyHaarCC)

wavethresh documentation built on Nov. 16, 2022, 5:16 p.m.