Calculating correlation divergence

Hi,

I know this topic became quite popular lately, but there is something I still cannot understand with currency correlation.
Every correlated pair (ie. EURUDS/USDCHF) has its own periods of divergence. How can this divergence be calculated? I need to know at a given point in time whether the pair is currently fully correlated, or in a middle of a divergence. How can I do that?

Thanks!

You can use the correl() function in excel on historical data and determine what is normal and what isn't.. orrrr you just open up the ratio chart of eurusd/usdchf aka eurchf and see if it's trading around its mean or not You pick the timeframe

Check out hedge trading thread by spieler

If you want to to compare an arry of X and Y data you'd have to use the Bravis Pearson function.
It's really simple. First calculate the covariance and varancie on it's own and put it in the equation.

http://en.wikipedia.org/wiki/Correlation

Thanks for all the answer, it is really helpful.
However, all these methods show how to calculate the general correlation between two currencies, and they are really effective with that. I can't find out how to check whether a given point in time is correlated or not.
I know the spieler system deals with that, but I just couldn't find the answer there. Perhaps I'm missing something?

memilavi, spieler calculates divergence from a reference point, the reference point being previous day's day close. to calculate divergence or convergence one must have a reference point

Could please you explain in a little more detail this ideea?
I am trying to write an algorithm for detecting divergence...