RE: st: programming question: obtaining statistics from clustered data

["Nick Winter" <nwinter@policystudies.com>]

-----Original Message-----

For the standard deviation the answer seems to be more difficult. At the

moment I only can think about a solution with a loop over the
observations 
within each cluster. There must be a better solution and I am sure that
I 
have overlooked somethink obvious. But anyway, you may use the following
as a 
starting point: 

>>>>>>>>>
You can use the alternate expression for variance to do the standard
deviation relatively straightforwardly:


gen X2=X^2
sort cluster
by cluster: gen sumX=sum(X)
by cluster: replace sumX=sumX[_N] - X
by cluster: gen sumX2=sum(X2)
by cluster: replace sumX2=sumX2[_N] - X2
by cluster: gen sd1 = sqrt(((_N-1)*sumX2 - sumX^2)/((_N-1)*(_N-2)))

--Nick Winter

<<winmail.dat>>