R help - Nov 2006 - Measuring the change in coupling of time series through time.

I am trying to work out what analysis to perform on two coupled
time-series. I want to ask how the coupling (i.e. lag or correlation)
changes over time.
 
The time series are of 4 replicate pred-prey time series where the
series appear more tightly coupled at peaks but drift apart during the
decline and trough phases of the cycles.
 
The time series are short (70-80 weeks/points replicated 4 times).
 
I have investigated cross-spectrum, cross correlation and cross-wavelet
techniques in R. But these either give the dominant lag, period or
correlation over time and wavelets appear to be the wrong technique to
ask what happens between peaks and troughs.
 
Has anyone else tried to ask this question and/or do they know of R
techniques to analyse this question?
 
Any ideas would be appreciated
 
Best wishes
 
Tom
 

........................................................................
............
Dr Tom C Cameron
Ecology and Evolution research group
IICB, University of Leeds
Leeds, UK
Office: +44 (0)113 343 2837
Lab:    +44 (0)113 343 2854
Fax:    +44 (0)113 343 2835


Email: t.c.cameron@leeds.ac.uk
Webpage: click here
<http://www.fbs.leeds.ac.uk/staff/profile.php?tag=Cameron_T> 

 

	[[alternative HTML version deleted]]