The square root of the second central momentum , characterizing the deviation of a given numerical dataset from its Mean Value :

\sigma(x) = \sqrt{\mu_2} = \sqrt{E[ ( x - \mu)^2 ]} = \sqrt{\frac{1}{N} \sum_{i=1}^N (x_i - \mu)^2}

where

	dataset length
	expectation operator

It is often used as normalization denominator for higher order central momentums : .

See also