How to efficiently calculate a running standard deviation

The answer is to use Welford's algorithm, which is very clearly defined after the "naive methods" in:

It's more numerically stable than either the two-pass or online simple sum of squares collectors suggested in other responses. The stability only really matters when you have lots of values that are close to each other as they lead to what is known as "catastrophic cancellation" in the floating point literature.

You might also want to brush up on the difference between dividing by the number of samples (N) and N-1 in the variance calculation (squared deviation). Dividing by N-1 leads to an unbiased estimate of variance from the sample, whereas dividing by N on average underestimates variance (because it doesn't take into account the variance between the sample mean and the true mean).

I wrote two blog entries on the topic which go into more details, including how to delete previous values online:

You can also take a look at my Java implement; the javadoc, source, and unit tests are all online:



Related Questions

How do I avoid rehashing overhead with std::set in multithreaded code?

How do I find elements with custom comparators with std::set for embedded targets?

How do I erase elements while iterating with std::set for embedded targets?

How do I provide stable iteration order with std::unordered_map for large datasets?

How do I reserve capacity ahead of time with std::unordered_map for large datasets?

How do I erase elements while iterating with std::unordered_map in multithreaded code?

How do I provide stable iteration order with std::map for embedded targets?

How do I provide stable iteration order with std::map in multithreaded code?

How do I avoid rehashing overhead with std::map in performance-sensitive code?

How do I merge two containers efficiently with std::map for embedded targets?