waveshrink
,
waveshrink.rts
or
waveshrink.ts
that are outcomes of WaveShrink procedure.
Objects of class
"waveshrink"
,
"waveshrink.rts"
or
"waveshrink.ts"
may be returned from calling function
waveshrink()
.
The classes
"waveshrink"
,
"waveshrink.rts"
and
"waveshrink.ts"
have methods for the following generic functions:
barplot, boxplot, eda.plot, stack.plot, summary
.
barplot
for each crystal, decomposes the energy into the energy attributable
to signal and residual.
boxplot
gives the boxplots of the
raw.dwt
object, see below for details.
eda.plot
has an important argument
residual
. If
residual=FALSE
,
eda.plot
consists of four subplots: a
stack.plot
, a
boxplot
, a plot of
raw.dwt
below
and a
barplot
; if
residual=TRUE
,
eda.plot
produces a residual plot,
an ACF plot, a
qqnorm
plot and a histogram with density estimate superimposed.
stack.plot
consists of three plots: data, signal and residual, and the sum of
the latter two yields data.
summary
contains the thresholds, the scale factors,
number of non-zero coefficients by crystal, Shapiro-Wilk normality test and
its P-value, Box-Ljung randomness test and its P-value.
The
"waveshrink"
object is a "smoothed" signal with the following attributes:
dwt
object; otherwise, it is the original input
x
.
raw.dwt
object.
raw.dwt
.
raw.dwt
.
M+2
, where
M
is the total number of crystals
in
raw.dwt
. The first
M
integers are the numbers of non-zero coefficients
in the crystals, and the last two integers are the degrees of freedom used
for computing the thresholds and the scale factors, respectively.