Hello everyone,
I have a panel dataset containing trade data for 85 countries, across 16 years with around 12-15 variables (continuous and indicator). I wanted to combine observations for 10 specific countries (sum continuous variables' data, take the max value of the indicator variables) so as to create an observation for the group. For example, let A, B, X and Y be the sample of countries. Let A and B be the countries for which I need to merge data to create a country M. Presently, I have data which looks like this:
This is just for one year, there is data for 15 more years.
I want it to look like this:
I experimented with collapse and egen, but couldn't figure out a way. Does anyone have any ideas about what I could try?
Many thanks,
Meemansa
I have a panel dataset containing trade data for 85 countries, across 16 years with around 12-15 variables (continuous and indicator). I wanted to combine observations for 10 specific countries (sum continuous variables' data, take the max value of the indicator variables) so as to create an observation for the group. For example, let A, B, X and Y be the sample of countries. Let A and B be the countries for which I need to merge data to create a country M. Presently, I have data which looks like this:
:
input year str2 exporter str2 importer exports contiguity 2000 "A" "B" 10000 0 2000 "A" "X" 12000 1 2000 "A" "Y" 15000 0 2000 "A" "A" . 0 2000 "B" "X" 45000 0 2000 "B" "Y" 34500 0 2000 "B" "A" 35600 0 2000 "B" "B" . 0 2000 "X" "A" 57600 1 2000 "X" "B" 25600 0 2000 "X" "X" . 0 2000 "X" "Y" 68700 0 2000 "Y" "X" 23400 0 2000 "Y" "Y" . 0 2000 "Y" "A" 78700 0 2000 "Y" "B" 76400 0 end
I want it to look like this:
| Year | Exporter | Importer | Export value | Contiguity |
| 2000 | M | X | sum of A to X & B to X (i.e. 12000+45000) | 1 |
| 2000 | M | Y | sum of A to Y & B to Y | 0 |
| 2000 | M | M | sum of A to B & B to A | 0 |
| 2000 | X | M | sum of X to A & X to B | 1 |
| 2000 | Y | M | sum of Y to A & Y to B | 0 |
Many thanks,
Meemansa