growth rate calculation

[Chad Kim]

Hello, I just started using stata in my university and this simple yet too hard for me right now problem is bothering me.
I have to compute gdp growth rate but instead this time the data was sorted in unfamiliar format

country                after   gdp_capita
"aaaaaa"                  0        2209
"aaaaaa"                  1        3885
"bbbbbb"                  0        3562
"bbbbbb"                  1        4459
"cccccc"                  0        2188

It was a lot easier when data was sorted horizontally. However this time I have no clue.

For example,

gen gdp_after = gdp_capita if after==1

gen gdp_before = gdp_capita if after==0

I tried this but it created missing values

 country                after   gdp_capita   gdp_after   gdp_before
"aaaaaa"                  0        2209                      2209 
"aaaaaa"                  1        3885         3885
"bbbbbb"                  0        3562                      3562
"bbbbbb"                  1        4459         4459
"cccccc"                  0        2188

Thus, I couldn't do

gen growth = (gdp_after - gdp_before)/gdp_before

It will be very much appreciated, if someone knows how to fix this.

Thank you

[William Lisowski]

Welcome to Statalist.

Perhaps this example will be useful to you.

* Example generated by -dataex-. To install: ssc install dataex
clear
input str8 country byte after int gdp_capita
"aaaaaa" 0 2209
"aaaaaa" 1 3885
"bbbbbb" 0 3562
"bbbbbb" 1 4459
end
generate growth = .
bysort country (after): replace growth = (gdp_capita-gdp_capita[_n-1])/gdp_capita[_n-1] if after
list, noobs sepby(country)

. list, noobs sepby(country)

  +---------------------------------------+
  | country   after   gdp_ca~a     growth |
  |---------------------------------------|
  |  aaaaaa       0       2209          . |
  |  aaaaaa       1       3885   .7587144 |
  |---------------------------------------|
  |  bbbbbb       0       3562          . |
  |  bbbbbb       1       4459   .2518248 |
  +---------------------------------------+

[Igor Paploski]

William's response is great because it does not require you to change between long and wide formats. However, if you decide you still would like to work with the data in a wide format, try:

* Example generated by -dataex-. To install: ssc install dataex
clear
input str3 country float(after gdp_capita)
"aaa" 0 2209
"aaa" 1 3885
"bbb" 0 3562
"bbb" 1 4459
"ccc" 0 2188
"ccc" 1 2321
end

reshape wide gdp_capita, i(country) j(after)