st: Computing GDP growth using a log-linear regression

[D Harwood <bus4dpah@leeds.ac.uk>]

Dear All,

I was wondering if anyone knew how to compute GDP growth rates by  
country. I have panel data arranged as follows:

Country             Year            GDP ($ 2000)
Benin               1997            1947193088
Benin               1998            2035610240
Benin               1999            2131227520
Benin               2000            2254838784
Benin               2001            2367580672
Benin               2002            2474121728
Benin               2003            2570612480
Benin               2004            2650301440
Benin               2005            2727160320
.
.
.

And so on for a number of countries. Arranged by country code and by year.

I found an example in the archives of how to calculate the growth  
rates (http://www.stata.com/statalist/archive/2006-05/msg00529.html)  
and yet when I try to do the same I get an error message 'Type  
mismatch', no matter which variable type I enter. I typed the  
following into the command box and came out with the output shown.

. egen minyearbycountry = min(year), by(code)
r; t=0.00 15:30:23

. gen yearcounter = year-minyearbycountry
r; t=0.01 15:30:31

. gen g = .
(288 missing values generated)
r; t=0.00 15:30:39

. levelsof code
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26  
27 28 29 30 31 32
r; t=0.01 15:30:48

. foreach var in `r(levels)' {
  2. reg lnGDP yearcounter if code =="`var'"
  3. replace g = _b[yearcounter] if code =="`var'"
  4. }
type mismatch
r(109); t=26.93 15:31:23

I would be grateful if anyone could point out where I have gone wrong?

Regards,

Dave


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/