Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: weighted mean, median and std error using egen

From   "Nick Cox" <>
To   <>
Subject   st: RE: weighted mean, median and std error using egen
Date   Tue, 20 Apr 2010 18:41:26 +0100

-egen- doesn't support weights as such. There are ways of subverting
that, which aren't visible in official -egen- functions, but even with
their aid, application of -egen- is unlikely to be as fast as using

As you are using pweights, the case is even more clear-cut. It really is
not a good idea to try some kind of back-door computations in svy
problems. You should use only procedures that fully understand -svy-.
The FAQ 

I want to calculate a variable containing weighted group summary
statistics, but I do not want to collapse the data and egen does not
support weights. How can I do this?

has _some_ relevance to your problem. 


Benjamin Villena Roldan

I need to construct mean, median, std.error variables for some groups of
the population using weights (pweights really). It seems to me that
Stata10 egen commands won't do this. Is there any way to do it by using
I know it is possible to do it by using summarize, but I think it will
take a much longer time. Am I right?

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index