Statalist


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

st: RE: Weighted daily mean


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Weighted daily mean
Date   Thu, 16 Oct 2008 12:17:45 +0100

This is not to me a simple question as you don't spell out fully what is
going on. 

I guess that you have a variable number of trades on each day. In that
situation, the mean over trades and the mean over days (the mean of
daily means) will probably differ. 

If you want a mean over trades, that would be easiest got by 

. summarize trade 

And if you want that in a variable (difficult to see why), you could
then go 

. gen trademeantot = r(mean) 

Your second statement is equivalent, but a bit inefficient. Still, by
the time you have typed two commands not one, you wouldn't notice. 

If you want a mean over days, each day being given equal weight, then
you could go

egen tag = tag(date) 
su trademean if tag 

and if necessary 

gen trademeantot = r(mean) 

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

Beatrice Crozza

maybe again a simple question, but I tried to read the help and I
didn't find a solution.

I have calculated a daily mean for my variable trade, typing:

egen trademean=mean(trade), by(date)

now I want to calculate the total mean, so I typed:

egne trademeantot=mean(trademean)

but I don't think that the result is correct.

Do you have any idea of how to calculate a mean, taking into account
the daily mean?


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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