Generating groups defined by multiple variables

[Tiffany Meier]

Hi Statalists,

I am trying to generate an enumeration variable for groups which are defined by other variables.
In the example below I likewise try to generate X which is distinct for every B sorted by A.

Code:

A   B   X
3   1   1
3   1   1
3   1   1
3   1   1
3   0   2
3   0   2   
3   0   2
7   1   3
7   1   3
7   1   3
7   1   3
7   0   4
7   0   4

Thank You

[Mauricio Caceres]

It depends. If you just want a unique ID, you can do "egen X = group(A B)" but if you want to replicate "X" then, since B is sorted in descending order, you'd have to do "gsort A -B, gen(X)"

[Tiffany Meier]

Thank you very much Mauricio!

[Marcos Almeida]

There is a trick issue here.

You decided to start the sequence with 1 for variable B, instead of 0. That can be confusing.

Ignoring this aspect and just giving a sequence may be a solution.

After all, there are levels of a categorical variable, and Stata will recognize which is which:

. egen float X2 = group(A B)

. list

     +----------------+
     | A   B   X   X2 |
     |----------------|
  1. | 3   1   1    2 |
  2. | 3   1   1    2 |
  3. | 3   1   1    2 |
  4. | 3   1   1    2 |
  5. | 3   0   2    1 |
     |----------------|
  6. | 3   0   2    1 |
  7. | 3   0   2    1 |
  8. | 7   1   3    4 |
  9. | 7   1   3    4 |
 10. | 7   1   3    4 |
     |----------------|
 11. | 7   1   3    4 |
 12. | 7   0   4    3 |
 13. | 7   0   4    3 |
     +----------------+

[Janine McMinn]

X