Posts tagged with "statistics"

Accurately computing running variance

The most obvious way to compute variance then would be to have two sums: one to accumulate the sum of the x's and another to accumulate the sums of the squares of the x's. If the x's are large and the differences between them small, direct evaluation of the equation above would require computing a small number as the difference of two large numbers, a red flag for numerical computing. The loss of precision can be so bad that the expression above evaluates to a negative number even though variance is always positive. See Comparing three methods of computing standard deviation for examples of just how bad the above formula can be.