Condensing observations

[Etsir Novich]

I have population data for a country per year. Currently, there is three year-observations per country which I would like to condense so that there is one row per country-year. Here is how the data looks like:

Year	Country Name	population level 1	population level 2	population level 3
1989	Italy	1117967	.	.
1989	Italy	.	1922792	.
1989	Italy	.	.	155041

As I have many countries in the dataset, I would like to automatize the process so I can get something like:

Year	Country Name	population level 1	population level 2	population level 3
1989	Italy	1117967	1922792	155041
1990	Italy	xxx	yyy	zzz
1991	Italy	mmm	nnn	ddd

I was approaching the problem from removing duplicates to "bysort-ing" and dropping observations perspective, but I cannot get it properly working.
Was having trouble tracking down specific guidance for this, so thank you so in advance for the help offered.

[Nick Cox]

You need something like

collapse (firstnm) ..., by(country year)

where the ... depend on variable names we can't see and any other details.

firstnm as a search term may catch other mentions.

[Etsir Novich]

Dear,
Thank you! Perfect solution.