Hi try this:
. sort cat
. by cat: egen sumcat=sum(cat)
. by cat: gen weight=cat/sumcat
. drop sumcat
Rajesh
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of FUKUGAWA, N.
Sent: 17 January 2007 05:48
To: [email protected]
Subject: st: how to create weight and aggregate data using the weight
dear all,
i need to aggregate the data using weight.
table 1
cat catid x
1 24 a
1 73 a
1 6 b
2 108 c
2 12 a
3 67 d
4 41 c
4 9 d
4 103 b
4 85 c
weight is inverse of the number of "catid" within "cat".
the number of "catid" can be computed as follows.
. bys cat:count
-> cat = 1
3
-> cat = 2
2
-> cat = 3
1
-> cat = 4
4
then table 1 with weight will be as follows.
table 2
cat catid x w
1 24 a .33
1 73 a .33
1 6 b .33
2 108 c .5
2 12 a .5
3 67 d 1
4 41 c .25
4 9 d .25
4 103 b .25
4 85 c .25
finally i need to aggregated the data by x.
table 3
x freq
a 1.166 (=.33+.33+.5)
b .58 (=.33+.25)
c 1 (=.5+.25+.25)
d 1.25 (=1+.25)
how can i obtain tables 2 and 3?
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/
