*egen pair_id = group(DIST).

[Daniel Gomez]

Hi,

I have created an id variable for a pair of countries(exporter importer) based on their distance (DIST):

egen pair_id = group(DIST).

However, in my base, the distance ISR-JOR is not exactly the same distance JOR-ISR considering all the decimals.

What can I do?


exporter importer pair_id year DIST
ISR JOR 2 1990 114.6373
ISR JOR 2 1998 114.6373
ISR JOR 2 1986 114.6373
ISR JOR 2 1994 114.6373
ISR JOR 2 2002 114.6373
ISR JOR 2 2006 114.6373
JOR ISR 3 2002 114.6373
JOR ISR 3 1998 114.6373
JOR ISR 3 2006 114.6373
JOR ISR 3 1990 114.6373
JOR ISR 3 1986 114.6373

Thanks!

[Wouter Wakker]

This is another way to form pair id's:

* Example generated by -dataex-. To install: ssc install dataex
clear
input str3(exporter importer) byte pair_id int year float DIST
"ISR" "JOR" 2 1990 114.6373
"ISR" "JOR" 2 1998 114.6373
"ISR" "JOR" 2 1986 114.6373
"ISR" "JOR" 2 1994 114.6373
"ISR" "JOR" 2 2002 114.6373
"ISR" "JOR" 2 2006 114.6373
"JOR" "ISR" 3 2002 114.6373
"JOR" "ISR" 3 1998 114.6373
"JOR" "ISR" 3 2006 114.6373
"JOR" "ISR" 3 1990 114.6373
"JOR" "ISR" 3 1986 114.6373
end

gen countrypair = cond(exporter < importer, exporter + " " + importer, importer + " " + exporter)
egen wanted = group(countrypair)
list, noobs abbr(14)

. list, noobs abbr(14)

  +------------------------------------------------------------------------+
  | exporter   importer   pair_id   year       DIST   countrypair   wanted |
  |------------------------------------------------------------------------|
  |      ISR        JOR         2   1990   114.6373       ISR JOR        1 |
  |      ISR        JOR         2   1998   114.6373       ISR JOR        1 |
  |      ISR        JOR         2   1986   114.6373       ISR JOR        1 |
  |      ISR        JOR         2   1994   114.6373       ISR JOR        1 |
  |      ISR        JOR         2   2002   114.6373       ISR JOR        1 |
  |------------------------------------------------------------------------|
  |      ISR        JOR         2   2006   114.6373       ISR JOR        1 |
  |      JOR        ISR         3   2002   114.6373       ISR JOR        1 |
  |      JOR        ISR         3   1998   114.6373       ISR JOR        1 |
  |      JOR        ISR         3   2006   114.6373       ISR JOR        1 |
  |      JOR        ISR         3   1990   114.6373       ISR JOR        1 |
  |------------------------------------------------------------------------|
  |      JOR        ISR         3   1986   114.6373       ISR JOR        1 |
  +------------------------------------------------------------------------+

[Daniel Gomez]

It works! Thank you!!!! Wouter Wakker

[Nick Cox]

See also https://www.stata-journal.com/articl...article=dm0043 for a broader discussion.

[Daniel Gomez]

Thank you Nick Cox !