Recode multiple observations to missing

[Carlos Ingeberg]

Hello Statalist

I need help on how i can recode multiple observations to missing. In the data example below welfare is a clone of var5, but i only want observations with certain values to be cloned. Also, i want the values in welfare to be set to missing in var5.

I am aware of the recode var5 (1/2=.), generate(welfare) command, however, this command only seems to work with two values. Thus, i cannot write recode var5 (1/2/3=.) as stata tells me "unknown el / in rule r(198);". It also doesnt set the values in welfare to missing in var5.


Any help is much appreciated!

* Example generated by -dataex-. For more info, type help dataex
clear
input byte(var5 welfare)
 6  6
 9  9
10 10
24 24
23 23
 4  4
18 18
15 15
18 18
 7  7
18 18
 7  7
 6  6
 4  4
10 10
11 11
 2  2
23 23
 5  5
 5  5
end

[Nick Cox]

Sorry, but what do you mean by "certain values"?

My guess is that you want to replace not recode here.

[Carlos Ingeberg]

Hi Nick, thanks for a quick reply!

By certain values i mean that i want observations with values 5, 7, 14, 15, 38, 40 and 42 to be cloned to welfare and set to missing in var5.

[Nick Cox]

Sorry, I am still guessing. here. Is this what you want? Or show enough technique?

* Example generated by -dataex-. For more info, type help dataex
clear
input byte var5
 6
 9
10
24
23
 4
18
15
18
 7
18
 7
 6
 4
10
11
 2
23
 5
 5
end
. 
. clonevar welfare = var5 if inlist(var5, 5, 7, 14, 15, 38, 40, 42)
. 
. list 

     +----------------+
     | var5   welfare |
     |----------------|
  1. |    6         . |
  2. |    9         . |
  3. |   10         . |
  4. |   24         . |
  5. |   23         . |
     |----------------|
  6. |    4         . |
  7. |   18         . |
  8. |   15        15 |
  9. |   18         . |
 10. |    7         7 |
     |----------------|
 11. |   18         . |
 12. |    7         7 |
 13. |    6         . |
 14. |    4         . |
 15. |   10         . |
     |----------------|
 16. |   11         . |
 17. |    2         . |
 18. |   23         . |
 19. |    5         5 |
 20. |    5         5 |
     +----------------+

. 
. replace var5 = . if !missing(welfare)

[Carlos Ingeberg]

This is exactly what i was looking for. Thank you so much!