Identifying and flagging matches

[Kate Clark]

I have a large dataset with rows that include a unique identifier 'id' and a unique time variable called 'timeto'. There are variable numbers of duplicate id's but all in the dataset are duplicated at one point or another. I would like to create a flag indicating when two rows with the same id have 'timeto' values that are <= 2. If there are more than one, I'd like to count up. Examples below:

starting data:

id timeto
1 3
2 7
2 8
2 0
2 14
2 15
3 4
3 6

ending data:
id timeto flag
1 3 0
2 7 1
2 8 1
2 0 0
2 14 2
2 15 2
3 4 1
3 6 1

Eventually, for those 'pairs' of rows with the same ID and timeto within 2, I'd be planning to pivot wide so there is a single row for each ID. Any help would be appreciated!

[Clyde Schechter]

What would you want to do if the data were like this?

id timeto
1 3
2 7
2 8
2 9
2 10
2 0
2 14
2 15
3 4
3 6

[Kate Clark]

What would you want to do if the data were like this?

id timeto
1 3
2 7
2 8
2 9
2 10
2 0
2 14
2 15
3 4
3 6

Hi! thanks for the response. In this case:

id timeto flag
1 3 0
2 7 1
2 8 1
2 9 1
2 10 1
2 0 0
2 14 2
2 15 2
3 4 1
3 6 1

I expect the situation with 4 values close to one another to be rare (the 2, 7-10 combo you included above) just based on the nature of the data, so if there were a rare occurrence I think i could figure it out more manually.

[Clyde Schechter]

* Example generated by -dataex-. For more info, type help dataex
clear
input byte(id timeto)
1  3
2  7
2  8
2  9
2 10
2  0
2 14
2 15
3  4
3  6
end

gen `c(obs_t)' obs_no = _n  
by id (obs_no), sort: gen byte close = inrange(abs(timeto[_n+1]-timeto), 0, 2) ///
    | inrange(abs(timeto-timeto[_n-1]), 0, 2)
by id (obs_no): gen flag = sum(close == 1 & (close[_n-1] == 0 | _n == 1))
replace flag = 0 if !close