Generate new variable without duplicates

[Deivyd VELASQUEZ]

Hello
I have one problem, I need to create a new variable:

One extrait of my dataset is the following:
id_observation: identification for each observation, there are duplicates
year:
country: 3 country
total observation for each country without discriminate the year

id_observation	year	country	total observations
1107	1996	A	71
1101	1993	A	71
1106	2003	A	71
1105	2007	A	71
1104	1987	A	71
1103	1989	A	71
1102	2007	A	71
1101	1993	A	71
2202	2013	B	114
2201	1975	B	114
2202	2013	B	114
2203	2009	B	114
2204	1993	B	114
2206	2011	B	114
2205	2014	B	114
3304	2002	C	612
3305	2009	C	612
3303	1997	C	612
3303	1997	C	612
3301	2013	C	612
3302	2007	C	612
3308	1981	C	612
3309	2004	C	612
3303	1997	C	612
3303	1997	C	612

I have the total of observations for each country, according to each year I'd like to create one new variable which tells me how many unique observations are in each one, without duplicates

Do you have any advices?

Thank you so much for your help

[Marcos Almeida]

You may think about using - duplicates tag - plus - generate() p for that matter.

[Deivyd VELASQUEZ]

You may think about using - duplicates tag - plus - generate() p for that matter.

I tried the first time:

duplicates tag id_observation year, gen(tag)

tab tag if country=="A"

And the total is the total observations with duplicates

For that reason, I don't know what to do

[Marcos Almeida]

If I understood right, with - tab tag - you’ll get the duplicates plus the unique values. The sum, logically, will be the total of observations. That being said, you may - keep - the unique values - or use the “if” clause to select your sample in each estimation. Hopefully that helps.

[Deivyd VELASQUEZ]

Thank you so much, it worked with if clause

[Marcos Almeida]

Thank you for informing your query reached a satisfactory reply.