| Download a pdf copy of this help file here |
The equation for the sample standard deviation (the unbiased estimate of the population
standard deviation when the mean is known), is:
(1)
Each xi is a sample from a Normal(m,s) distribution, so Equation (1) can be rewritten:
(2)
A Chisq(n) distribution is the sum of n squared unit Normal distributions:
![]()
so Equation (2) becomes:
(3)
Using the pivotal method, we can rearrange Equation (3) to provide an estimate of s :

There is some possibility of getting
mixed
up with the sample standard deviation when the mean is unknown, which
has a denominator of (n-1), so the equation is better written:

This last equation can be directly used in an
ModelRisk model to simulate
uncertainty about s.