Summing variable across different units by year

[Chinmay Sharma]

Hi All,

I have data that looks like the following:

* Example generated by -dataex-. To install: ssc install dataex
clear
input float(country year population)
1 1980 100
1 1981 150
1 1982 200
2 1980 200
2 1981 300
2 1982 300
end

------------------ copy up to and including the previous line ------------------


In the above, 1 and 2 represent country identifiers. What I wish to do, is for each year (1980, 1981 etc), I wish to create a new variable, which represents the sum of country 1 and 2's population. In particular, the final dataset I envision should looks something like:

* Example generated by -dataex-. To install: ssc install dataex
clear
input float(var1 var2 var3)
1 1980 100
1 1981 150
1 1982 200
2 1980 200
2 1981 300
2 1982 300
3 1980 300
3 1981 450
3 1982 500
end

In the above, 3 is a new identifier representing the combination of country 1 and 2. Reshaping the data and summing across rows would work, but that might be overkill for this particular operation. Any suggestions are most welcome!

Best,
CS

[Nick Cox]

help egen

egen wanted = total(pop), by(country)

[Chinmay Sharma]

Thanks a lot Nick Cox .