spec.taper(x, p=0.1)
0 and
0.5 giving the fraction of the data to be tapered
on each end of the series.
In the case of a multivariate time series,
p can either be a scalar
or a vector with as many elements as columns in the series, each element
of the vector
p gives the proportion of the data to be tapered
in the respective column.
x containing the result of tapering the input
x.
Values of the result that are inside the tapering window will be smaller in
absolute value than the corresponding values in the input.
Implements a split cosine bell taper. Let
p be the portion to be tapered
at each end of the series by
n the length of the series, then for
m=np
the split cosine bell taper is
Bloomfield, P. (1976).
Fourier Analysis of Time Series: An Introduction.
Wiley, New York.
The chapter "Analyzing Time Series" of the S-PLUS Guide to Statistical and Mathematical Analysis.
# taper 10% of each end of the housing start data hstaper <- spec.taper(hstart) hstart3 <- tsmatrix(hstart, hstart, hstart) # replicate hstart three times # taper 10%, 20% and 40% of each end hstaper3 <- spec.taper(hstart3, c(.1, .2, .4)) # the split-cosine-bell window for p=.25 tsplot(spec.taper(rep(1, 100), p=.25))