Generating a variable dependent on the year

[Surya Singh]

Hi,
I am trying to generate a variable (i.e. labor income_breastfeeding; employmentstatus_breastfeeding; etc...) for each of my observations in respect to the year that the mother was breastfeeding (year_breastfeeding). My data is panel data set with child-mother pairs and for example looks like this:

ID_child ID_mother year labor income year_breastfeeding laborincome_breastfeeding (desired var with desired values)
1 a 1989 34568 1990 36789
1 a 1990 36789 1990 36789
1 a 1991 37329 1990 36789
2 b 1983 50681 1984 54808
2 b 1984 54808 1984 54808
2 b 1985 55029 1984 54808
2 b 1886 56928 1984 54808

Therefore, the laborincome_breastfeeding for ID_child 1, ID_mother a, should be 36789 since 1990 is the year_breastfeeding and for ID_child 2, ID_mother b, laborincome_breastfeeding should be 54808.
I have tried the egen command; however, cannot seem to figure out how to do it correctly. I'm a bit new at state but perhaps a loop over observations would be best?

Any help would be greatly appreciated, thank you for your time and help!

Regards,
Surya

[Nick Cox]

Your variables seem a bit tangled up, but the spirit of what you want seems to be

. clear 

. input ID_child str1 ID_mother year1 income year2  wanted

      ID_child  ID_mother      year1     income      year2     wanted
  1. 1 a 1989 34568 1990 36789
  2. 1 a 1990 36789 1990 36789
  3. 1 a 1991 37329 1990 36789
  4. 2 b 1983 50681 1984 54808
  5. 2 b 1984 54808 1984 54808
  6. 2 b 1985 55029 1984 54808
  7. 2 b 1986 56928 1984 54808 
  8. end 

. 
. egen income_breastfeeding = total(income * (year1 == year2)), by(ID_child ID_mother)  

. assert wanted == income_breastfeeding 

. 
. l, sepby(ID*) 

     +------------------------------------------------------------------+
     | ID_child   ID_mot~r   year1   income   year2   wanted   income~g |
     |------------------------------------------------------------------|
  1. |        1          a    1989    34568    1990    36789      36789 |
  2. |        1          a    1990    36789    1990    36789      36789 |
  3. |        1          a    1991    37329    1990    36789      36789 |
     |------------------------------------------------------------------|
  4. |        2          b    1983    50681    1984    54808      54808 |
  5. |        2          b    1984    54808    1984    54808      54808 |
  6. |        2          b    1985    55029    1984    54808      54808 |
  7. |        2          b    1986    56928    1984    54808      54808 |
     +------------------------------------------------------------------+

[Surya Singh]

Thank you Nick!

Your variables seem a bit tangled up, but the spirit of what you want seems to be

. clear

. input ID_child str1 ID_mother year1 income year2 wanted

ID_child ID_mother year1 income year2 wanted
1. 1 a 1989 34568 1990 36789
2. 1 a 1990 36789 1990 36789
3. 1 a 1991 37329 1990 36789
4. 2 b 1983 50681 1984 54808
5. 2 b 1984 54808 1984 54808
6. 2 b 1985 55029 1984 54808
7. 2 b 1986 56928 1984 54808
8. end

.
. egen income_breastfeeding = total(income * (year1 == year2)), by(ID_child ID_mother)

. assert wanted == income_breastfeeding

.
. l, sepby(ID*)

+------------------------------------------------------------------+
| ID_child ID_mot~r year1 income year2 wanted income~g |
|------------------------------------------------------------------|
1. | 1 a 1989 34568 1990 36789 36789 |
2. | 1 a 1990 36789 1990 36789 36789 |
3. | 1 a 1991 37329 1990 36789 36789 |
|------------------------------------------------------------------|
4. | 2 b 1983 50681 1984 54808 54808 |
5. | 2 b 1984 54808 1984 54808 54808 |
6. | 2 b 1985 55029 1984 54808 54808 |
7. | 2 b 1986 56928 1984 54808 54808 |
+------------------------------------------------------------------+