Creating variable for other events in same region/year

[Birte Olsen]

Hi everyone!

I am dealing with a worldwide data set on conflict events with an unbalanced panel structure, where I am given the variable "id", "year", as well as "region" and "location".

I want to create a binary variable "other_conflicts_region" that identifies whether there were other conflicts in the same region in the same year. I also want to create a binary variable "other_conflicts_world" that identifies whether there were any other conflicts in the world in the same year.

Here is a subset of my data:

* Example generated by -dataex-. For more info, type help dataex
clear
input str8 id double year str8 region str28 location
"100_1945" 1945 "Asia"     "Vietnam"                     
"101_1945" 1945 "Asia"     "Indonesia"                   
"238_1945" 1945 "Europe"   "USSR"                        
"129_1946" 1946 "Americas" "Colombia"                    
"376_1946" 1946 "Europe"   "Estonia"                     
"380_1946" 1946 "Europe"   "Latvia"                      
"381_1946" 1946 "Europe"   "Lithuania"                   
"153_1946" 1946 "Africa"   "Nigeria"                     
"199_1946" 1946 "Africa"   "Eritrea"                     
"378_1946" 1946 "Asia"     "Iran"                        
"109_1946" 1946 "Asia"     "Palestine"                   
"425_1946" 1946 "Asia"     "China"                       
"100_1946" 1946 "Asia"     "Vietnam"                     
"91_1946"  1946 "Asia"     "Philippines"                 
"101_1946" 1946 "Asia"     "Indonesia"                   
"238_1946" 1946 "Europe"   "USSR"                        
"432_1946" 1946 "Americas" "Bolivia"                     
"371_1946" 1946 "Asia"     "Cambodia (Kampuchea)"        
"368_1946" 1946 "Asia"     "Princely State of Travencore"
"364_1946" 1946 "Asia"     "Laos"                        
"401_1946" 1946 "Europe"   "Greece"                      
"129_1947" 1947 "Americas" "Colombia"                    
"166_1947" 1947 "Americas" "Paraguay"
                                   
end

I suspect I need to build my code with by and sort, but I am pretty new to this, so I would really appreciate your help and thoughts to get me started!

Kind regards,
Birte
(Stata 17.0 SE)

[Nick Cox]

The number of other people in my group is (the number of people in my group) MINUS 1 (that's me). Similarly,

.

 bysort region year : gen others = _N - 1

. gen wanted = others > 0  

. list if year == 1946, sepby(region)

     +-----------------------------------------------------------------------------+
     |       id   year     region                       location   others   wanted |
     |-----------------------------------------------------------------------------|
  1. | 199_1946   1946     Africa                        Eritrea        1        1 |
  2. | 153_1946   1946     Africa                        Nigeria        1        1 |
     |-----------------------------------------------------------------------------|
  3. | 129_1946   1946   Americas                       Colombia        1        1 |
  4. | 432_1946   1946   Americas                        Bolivia        1        1 |
     |-----------------------------------------------------------------------------|
  9. | 109_1946   1946       Asia                      Palestine        8        1 |
 10. | 368_1946   1946       Asia   Princely State of Travencore        8        1 |
 11. | 101_1946   1946       Asia                      Indonesia        8        1 |
 12. | 378_1946   1946       Asia                           Iran        8        1 |
 13. | 100_1946   1946       Asia                        Vietnam        8        1 |
 14. |  91_1946   1946       Asia                    Philippines        8        1 |
 15. | 364_1946   1946       Asia                           Laos        8        1 |
 16. | 425_1946   1946       Asia                          China        8        1 |
 17. | 371_1946   1946       Asia           Cambodia (Kampuchea)        8        1 |
     |-----------------------------------------------------------------------------|
 19. | 381_1946   1946     Europe                      Lithuania        4        1 |
 20. | 376_1946   1946     Europe                        Estonia        4        1 |
 21. | 401_1946   1946     Europe                         Greece        4        1 |
 22. | 238_1946   1946     Europe                           USSR        4        1 |
 23. | 380_1946   1946     Europe                         Latvia        4        1 |
     +-----------------------------------------------------------------------------+

I went around the block with some other command until I re-discovered this simpler way to think about it.

[Birte Olsen]

Thank you Nick - that worked perfectly and was very intuitive to understand!