PDA

View Full Version : Is there a way to compute a running correlation coefficient?



blake82
March 14th, 2009, 07:17 PM
I'm trying to find a way to compute a correlation coefficient on a growing data set, without having to load the entire data set to recalculate each time an element is added.

For instance, this is possible with mean and standard deviation by keeping a sum of the values (and a sum of the square of the values for stDev).

Does anyone know of a method?

cmat
March 14th, 2009, 07:20 PM
Yes you can. What programming language are you using?

blake82
March 14th, 2009, 07:21 PM
Excellent!

I'm using python