Prev: books and movies
Next: CSSM Chronology in Google Groups
From: sanam on 22 Mar 2010 10:24
Hi. What are the alternatives we have to cross-correllation and auto-correllation??
If I want to find degree to similarity between two sampled signals what options do I have besides doing the obvious correlation??

Thanks in advance.
From: Wayne King on 22 Mar 2010 18:50
sanam <sanamsingh(a)hotmail.com> wrote in message <713567448.421006.1269282290624.JavaMail.root(a)gallium.mathforum.org>...
> Hi. What are the alternatives we have to cross-correllation and auto-correllation??
> If I want to find degree to similarity between two sampled signals what options do I have besides doing the obvious correlation??
>
> Thanks in advance.

There are a number of options. There is coherence, which is basically frequency-domain correlation. There is also the information theoretic concept of mutual information.

Wayne
From: Manthos Vogiatzoglou on 22 Mar 2010 19:27
From statistical - theoretical point you can look at Kendal's tau or Spearman's rho or maybe copulas. For theory on copulas and what they do you can look the documentation in the statistics toolbox or for research papers fron AJ PAtton, Genest, Chen and Fan, Joe, McNeil and others
 | 
Pages: 1
Prev: books and movies
Next: CSSM Chronology in Google Groups