About Pymutt

Pymutt is an implementation of Thomson's (1982) multi-taper fourier spectral estimator plus a python interface. The core code is due to Lees and Park (1995) and uses the conventions of Percival and Walden (1993).

Requirements

• Building pymutt requires the python-dev package.
• Using pymutt requires numpy.
• Most of the examples also require matplotlib.

Installation

Pymutt builds and installs properly on * win7-64 using 32-bit python * linux-64 using 64-bit python It almost surely installs properly on * linux-32 and * win-32. It currently fails on win7-64 using 64-bit python.

To install pymutt execute

python setup.py build

and as root or administrator

python setup.py install

On windows I used mingw and the mingw shell.

Usage

Go to the examples subdirectory and execute

python runme.py

and it will print out help for runme.py as well as the self docs for pymutt. To run a test case try

python runme.py -N 3

and you should get a figure closely resembling Percival and Walden figure 512.

Look at the code in examples/minimalexamples.py for some easy-to-copy examples.

Other Code

For a versatile, extensive, and newer fortran system for multi-taper fourier transforms see

http://uniandes.academia.edu/gprieto

For a python interface to the preceding see

https://svn.geophysik.uni-muenchen.de/trac/mtspecpy/wiki

Also see http://nipy.org/nitime/ for an extensive time-series analysis package including multi-taper spectral estimation.

References

Lees, J.M., and J. Parks (1995), Multi-taper Spectral Analysis: A

Stand-alone C Subroutine, Computers and Geology, 21(2), 199-236. Percival, D.B., and A.T. Walden (1993) Spectral Analysis for Physical Applications, Cambridge University Press, Cambridge. Thomson, D.J. (1982), Spectral Estimation and Harmonic Analysis, IEEE Proc., 70, 1055-1096.