psd: Adaptive, sine-multitaper power spectral density estimation

Power spectral density estimates are produced through iterative refinement of the optimal number of sine-tapers at each frequency. The optimization procedure is based on the method of Riedel and Sidorenko (1995), which applies smoothing that varies with frequency to minimize the sum of variance and bias at each point.

Version: 0.4-1
Depends: R (≥ 2.14.1), stats, utils, graphics, grDevices, fftw (≥ 1.0-3)
Imports: RColorBrewer, signal, zoo
Suggests: bspec, ggplot2 (≥ 0.9), knitr, multitaper, plyr, RSEIS, rbenchmark, reshape2
Published: 2014-04-16
Author: Robert L. Parker and Andrew J. Barbour
Maintainer: Andrew J. Barbour <andy.barbour at>
License: GPL-2 | GPL-3 [expanded from: GPL]
URL: and for the citation.
NeedsCompilation: yes
Citation: psd citation info
Materials: NEWS
In views: TimeSeries
CRAN checks: psd results


Reference manual: psd.pdf
Vignettes: DFT benchmarks: fft vs FFT.
Normalization of power spectral density estimates.
An overview of psd.
Package source: psd_0.4-1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X Snow Leopard binaries: r-release: psd_0.4-1.tgz, r-oldrel: psd_0.4-1.tgz
OS X Mavericks binaries: r-release: not available
Old sources: psd archive

Reverse dependencies:

Reverse suggests: multitaper