locits (version 1.4)

HwdS: Compute the non-decimated Haar wavelet transform without using periodic boundary conditions.

Description

Function uses the filter function to achieve its aims.

Usage

HwdS(x)

Arguments

x

A vector of dyadic length that you wish to transform.

Value

• An object of class wd which contains the nondecimated Haar transform of the input series, x without periodic boundary conditions (nor interval, nor reflection).

Details

The regular wd function that can compute the non-decimated transform uses different kinds of boundary conditions, which can result in coefficients being used multiply for consideration in a test of stationarity, and distort results. This function only computes Haar coefficients on the data it can, without wrapround.

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.

See Also

ewspecHaarNonPer, getridofendNA

Examples

#
# Apply Haar transform to Gaussian data
#
HwdS(rnorm(32))
#Class 'wd' : Discrete Wavelet Transform Object:
#       ~~  : List with 8 components with names
#              C D nlevels fl.dbase filter type bc date 
#
#$C and $D are LONG coefficient vectors
#
#Created on : Tue Jul 17 15:14:59 2012 
#Type of decomposition:  station 
#
#summary(.):
#----------
#Levels:  5 
#Length of original:  32 
#Filter was:  Haar wavelet 
#Boundary handling:  periodic 
#Transform type:  station 
#Date:  Tue Jul 17 15:14:59 2012