mvcwt-package: Wavelet transform of multivariate time series

In mvcwt: Wavelet Analysis of Multiple Time Series

Description

Computes continuous wavelet transform of multiple irregularly sampled time series.

Details

Package:	mvcwt
Type:	Package
Version:	1.3
Date:	2013-10-27
License:	GPL

The main functions are mvcwt, which computes the wavelet transform of multiple time series, and wmr, which computes the wavelet modulus ratio, a measure of time series coherence.

Note that this is a complete rewrite of the code used in the reference below, and as such it is not well tested. It may give different or inaccurate results. I recommend you run tests on known data.

The most recent development version of this package can be found at https://github.com/thk686/mvcwt.

Author(s)

Timothy H. Keitt (http://www.keittlab.org)

Tim Keitt <[e-mail]>

References

Keitt, T. H. 2008. Coherent ecological dynamics induced by large-scale disturbance. Nature 454:331-4. doi:10.1038/nature06935.

Examples

## Not run: 
  x = seq(-pi, pi, len = 200)
  y1 = sin(8 * x) + sin(32 * x)
  y2 = sin(8 * (x + pi/8)) + sin(32 * x)
  matplot(x, cbind(y1, y2), type = "l", lty = 1)
  w = mvcwt(x, cbind(y1, y2))
  plot(w, var = 1:2, scale = 2^seq(log2(min(w$y)), log2(max(w$y)), len = 5))
  mr = wmr(w, smoothing = 2)
  image(mr, reset.par = FALSE)
  contour(mr, levels = c(0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99), add = TRUE)
## End(Not run)

Example output

Warning message:
executing %dopar% sequentially: no parallel backend registered 

mvcwt documentation built on May 2, 2019, 1:59 p.m.