Software: Analysis of univariate and bivariate time series

[Brandon Whitcher (whitcher@eurandom.tue.nl)]

#8 Software: Analysis of univariate and bivariate time series

WaveCov: An S-Plus software package for the analysis of univariate
and bivariate time series using wavelets

Author: Brandon Whitcher
whitcher@eurandom.tue.nl
http://www.eurandom.tue.nl/whitcher

We introduce a suite of S-Plus functions and C code in order to
perform a wavelet analysis of covariance and correlation between two
time series (Whitcher 1998). The necessary functions are described and
used to analyze the recorded height of ocean waves measured by two
different instruments in Cape Henry near Virginia Beach, Virginia.
The bivariate spectral analysis of Percival (1994) is also provided in
order to compare and contrast the wavelet methodology with classical
Fourier techniques.

Software is available for download from StatLib

http://lib.stat.cmu.edu/S/wavecov

and EURANDOM

http://www.eurandom.tue.nl/whitcher/papers.html

References

Percival, D. B. (1994). Spectral analysis of univariate and bivariate time
series. In J. L. Stanford and S. B. Vardeman (Eds.), <EM>Statistical
Methods for Physical Science</EM>, Volume 28 of <EM>Methods of
Experimental Physics</EM>, pp. 313-348. Boston: Academic Press, Inc.

Whitcher, B. (1998). <EM>Assessing Nonstationary Time Series Using
Wavelets</EM>. Ph. D. thesis, University of Washington.