Welcome to MathOverflow

Question

3

What is the best algorithm for computing covariance that would be accurate for a large number of values like 100,000 or more?

flag	asked Nov 18 2009 at 20:19 Kim Greene 1,228●3●20●36

One quick note. I know some people will think this is basic but I have looked in a lot of places (numerical analysis texts, computational statistics texts and google scholar) and I've never seen it discussed. Doing the straightforward thing does not work for large datasets. – Kim Greene Nov 18 2009 at 20:26

John D. Cook · Accepted Answer · 2009-11-18 20:44:43Z UTC

3

Check out How to calculate correlation accurately. There are two common formulas that are algebraically equivalent but one has much better numerical properties than the other.

answered Nov 18 2009 at 20:44

John D. Cook
2,029●16●37

1

Yes, but that is only a partial answer. The problem is that the “good” formula requires two passes over the data, one to compute the arithmetic means and one to compute the quadratic sums. It would be very handy to have an algorithm that uses only a single pass, yet is numerically stable. Surely, someone must have thought of one? It doesn't seem to me an intractable problem. – Harald Hanche-Olsen Nov 18 2009 at 21:53

1

Here's a way to compute standard deviation in one pass. I imagine you could take the ideas from this algorithm to create an accurate single-pass algorithm for covariance. johndcook.com/standard_deviation.html – John D. Cook Nov 19 2009 at 0:17

Mike B · Answer 2 · 2009-11-19 19:26:02Z UTC

3

The single-pass and parallel versions at Wikipedia may be what you're looking for. The single pass version is more numerically stable, but moves a division into the inner loop, which may hurt performance.

answered Nov 19 2009 at 19:26

Mike B
31●1

What’s an efficient way to calculate covariance for a large data set?

Remember to vote up questions/answers you find interesting or helpful (requires 15 reputation points)

2 Answers

You can accept an answer to one of your own questions by clicking the check mark next to it. This awards 15 reputation points to the person who answered and 2 reputation points to you.