Create average variable

[budi doremi]

Hi there..

Im a new user of stata. I need a basic help. I have a panel data. The structure for example

Region Year Income
A 2009 4.000
2010 5.000
2011 6.000
B 2009 3.000
2010 2.500
2011 7.200
Help me please to create average variabel of A n B.

Appreciate anyone's helps

[Carlo Lazzaro]

Budi:
you may want something along the following lines:

use "https://www.stata-press.com/data/r17/nlswork.dta"
bysort idcode (year): egen wanted=mean(ln_wage)
list idcode year ln_wage wanted if idcode<=2
       +-------------------------------------+
       | idcode   year    ln_wage     wanted |
       |-------------------------------------|
    1. |      1     70   1.451214   2.040433 |
    2. |      1     71    1.02862   2.040433 |
    3. |      1     72   1.589977   2.040433 |
    4. |      1     73   1.780273   2.040433 |
    5. |      1     75   1.777012   2.040433 |
       |-------------------------------------|
    6. |      1     77   1.778681   2.040433 |
    7. |      1     78   2.493976   2.040433 |
    8. |      1     80   2.551715   2.040433 |
    9. |      1     83   2.420261   2.040433 |
   10. |      1     85   2.614172   2.040433 |
       |-------------------------------------|
   11. |      1     87   2.536374   2.040433 |
   12. |      1     88   2.462927   2.040433 |
   13. |      2     71   1.360348   1.688055 |
   14. |      2     72   1.206198   1.688055 |
   15. |      2     73   1.549883   1.688055 |
       |-------------------------------------|
   16. |      2     75   1.832581   1.688055 |
   17. |      2     77   1.726721   1.688055 |
   18. |      2     78    1.68991   1.688055 |
   19. |      2     80   1.726964   1.688055 |
   20. |      2     82   1.808289   1.688055 |
       |-------------------------------------|
   21. |      2     83   1.863417   1.688055 |
   22. |      2     85   1.789367   1.688055 |
   23. |      2     87    1.84653   1.688055 |
   24. |      2     88   1.856449   1.688055 |
       +-------------------------------------+

.

[Nick Cox]

If your data really are like this

* Example generated by -dataex-. For more info, type help dataex
clear
input str12 region double income int year
"A"   4 2009
""    5 2010
""    6 2011
"B"   3 2009
""  2.5 2010
""  7.2 2011
end

Then you'll need to fill in missing values:

replace region = region[_n-1] if missing(region)

as explained at https://www.stata.com/support/faqs/d...issing-values/

The mean of ln wage as in @Carlo Lazzaro's helpful answer is the log of the geometric mean, and the geometric mean is often a better summary measure than the mean for incomes.