[R-sig-eco] Multivariate cross-correlation of time series

[Gavin Simpson]

On Thu, 2010-08-12 at 04:26 -0700, Dragos Zaharescu wrote:
> Hi everyone,
> 
> Sorry for cross-posting.I would like to ask if anyone knows of a R function 
> that performs multivariate corss-correlation (temporal series).
> I amconducting an alaysis of climate change effects on lake bed core variables. 
> All are time series. Likewise, I would like to infer the significance of the 
> correlation coefficients by permutation test.
> Many thanks in advance,

I'm not aware of anything for multivariate cross correlation, but do
check the TimeSeries task view on CRAN.

If the sediment core variables are few, then you could (i) treat the
variables independently one by one in turn using univariate approach, or
(ii) reduce them to a small set of ordination axes and compute
univariately as (i).

I did something similar to (ii) in a recent paper in Limnology and
Oceanography [1], although I was never happy with the PCA part of this.

Whatever you do, I think you'll end up needing to fit some form of model
and account for temporal autocorrelation. The permutation approach will
require producing a set of restricted permutations (see permute package
on R-forge, in the vegan project, for a preliminary stab at this).

Hope some of this helps.

If this is sediment core work, as I too am a palaeolimnologist, perhaps
we could chat off-list if you require further assistance or want to talk
about the approach I used in the above paper.

All the best,

G
[1] http://www.aslo.org/lo/toc/vol_54/issue_6_part_2/2529.pdf

> 
> Dragos
> 
> 
> Dragos Zaharescu
> Faculty of Biological Sciences
> Vigo University, apd. 137
> 36310, Vigo, SPAIN
> zaha_dragos at yahoo.com
> zdragos at uvigo.es
> http://webs.uvigo.es/zdragos/
> 
> 
>       
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology

-- 
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
 Dr. Gavin Simpson             [t] +44 (0)20 7679 0522
 ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
 Pearson Building,             [e] gavin.simpsonATNOSPAMucl.ac.uk
 Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
 UK. WC1E 6BT.                 [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%