st: data manipulation: pairs of variables

[Kit Baum <baum@bc.edu>]

Falko said

Please consider a stylized version of my dataset:

var1	var2	var3
1		2		223
1		3		12
2		1		7544
2		3		7864
3		1		434
3		2		12334

You can think of var1 and var2 as identifying different regions in a
country and var3 could measure migration between region pairs, for 
example
223 persons migrate from 1 to 2. There are three regions. Migrating 
within
the same region is not possible, that is why there are no (1,1),(2,2) or
(3,3) observations for var1 and var2.
...
My plan is to use var4 to calculate the net migration volume between
regions, which means I want to calculate:
223-7544
12-434
7864-12334


clear
input var1 var2 var3
1		2		223
1		3		12
2		1		7544
2		3		7864
3		1		434
3		2		12334
end
g reg1 = cond(var1>var2,var1,var2)
g reg2 = min(var1,var2)
replace var3 = cond(var1>var2,-var3,var3)
collapse (sum) var3, by(reg1 reg2)
l


Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html

*
*   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/