artlev {adlift}R Documentation

artlev

Description

This function splits the coefficients into levels according to increasing quantiles of the removed interval lengths.

Usage

artlev(y, rem)

Arguments

y

a vector of the removed interval lengths (in the order of removelist).

rem

vector of indices of the removed points (from the output of the forward transform).

Details

The function finds the median of the removed interval lengths, and takes all pointsin indices with removed interval lengths at most this value as the first artificial level. These indices are now not considered in later groups. The cut-off value, q, is now increased to the 75th percentile, and the indices at most this value are grouped into the second level. The procedure is continued with successive percentiles (1+q)/2 until all indices are grouped. At each stage, the level size is checked to ensure it has at least 10 elements, and if not, the level is taken together with the next level (i.e. the present percentile is ignored, and increased to the q value).

Value

p

a list of the grouped indices of removelist (in decreasing group size) indicating thresholding groups.

Author(s)

Matt Nunes (nunesrpackages@gmail.com), Marina Knight

See Also

denoise,

Examples

#create test signal data
#
x<-runif(100)
y<-make.signal2("blocks",x=x)
#
#perform forward transform...
#
out<-fwtnp(x,y,LocalPred=AdaptNeigh,neighbours=2)
#
al<-artlev(out$lengthsremove,out$removelist)
#
#
# the indices of removelist split into levels:
al
#

[Package adlift version 1.4-5 Index]