Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: AW: How to compute sigma or sum


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: AW: How to compute sigma or sum
Date   Sun, 3 Feb 2008 17:04:02 -0000

It seems that this doesn't answer the original question, and I am not
sure what does. But as -summarize- has been mentioned, correctly, as a
good way to get the sum of a variable, let me emphasise that if
efficiency is of concern, then -summarize, meanonly- is more efficient. 

The StataCorp option name -meanonly- may well mislead here, as the
option in question does more than calculate means. If more detail is
needed, see 

SJ-7-3  st0135  Stata tip 50: Efficient use of summarize

Q3/07   SJ 7(3):438--439           

tip on using the meanonly option of summarize to more
quickly determine the mean and other statistics of a
large dataset. 

Also in what follows, Sergiy has 

local sigma=r(sum)
display `sigma'

There's no need for the indirection for immediate display, as 

display r(sum) 

will do it in one. For the later use which Sergiy has in mind, 
storage naturally is important, but a scalar is more precise than a
local.  

tempname sigma 
scalar `sigma' = r(sum) 

Nick 
n.j.cox@durham.ac.uk 

Sergiy Radyakin

Martin Weiss has shown an example where a new variable is created and
its values for all observations are set to the sum of the values of
another variable. Although indeed this provides the desired sum, it is
rather inefficient in terms of both memory use and computational
performance.

To find a sum of the values of a variable, use

summarize x
display r(sum)

If need to reuse this value later in your program, store it in a local:
local sigma=r(sum)
display `sigma'

A number of other related statistics (min, max, sd) can be obtained in
a similar fashion. Type

return list

after summarize (or any other command) to see the results available.


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index