Hi,
One way to do it would be.
. collapse (sum) weight, by(x)
However, this will alter the dataset in mem. So type preserve first.
You could also try
. bysort x: egen sumweight=sum(weight)
This creates a new variable sumweight with the sum of the weights..
Hope this helps
rajesh
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of FUKUGAWA, N.
Sent: 16 January 2007 11:42
To: [email protected]
Subject: st: "table" command with weight
dear all,
i need to aggregate the data using weight.
the original data
cat catid x weight
1 1 a .33
1 2 a .33
1 3 b .33
2 1 c .5
2 2 a .5
3 1 d 1
4 1 c .25
4 2 d .25
4 3 b .25
4 1 c .25
the aggregated data
x freq
a 1.166(=.33+.33+.5)
b .58(=.33+.25)
c 1(=.5+.25+.25)
d 1.25(=1+.25)
please note i need to create weight (the proportion within the
category) because it does not exist in the original data.
this is to avoid double counting.
the number of rows is 9m.
the number of variables is 100.
thank you in advance.
--
FUKUGAWA Nobuya, Ph.D.
[email protected]
*
* 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/
*
* 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/
