Generating dummy variable using a range of dates

[Kanishk Kapur]

Hi,

I have a dataset of president's and their approval ratings. I would like to control for the entity fixed effects of every administration when I analyze the data. So, I need to divide the data into different administrations using a dummy variable. For example, President Obama's approval ratings need to be divided between his 2 terms: the 1st one from 2009q1 to 2012q4 and the 2nd one from 2013q1 to 2016q4. My date variable is in the format %tq (displayed as 2008q1). I am trying to create a dummy variable that is equal to 1 for the duration of every president's term.

Thanks a lot,
Kanishk

[Josh Clothiaux]

Hi,

I assume that the maximum number of quarters per term that a President holds office is 16.

bysort president_var (timeperiod): gen first_term = _n < = 16
gen second_term = 1 - first_term

[Nick Cox]

All quarterly dates that start 4 year terms have remainder 4 on division by 16:

. mata : mod(yq((2001, 2005, 2009, 2013), (1,1,1,1)), 16)
       1   2   3   4
    +-----------------+
  1 |  4   4   4   4  |
    +-----------------+

Therefore

 
. webuse turksales, clear

. gen spell = sum(mod(t, 16) == 4)

. tab spell, gen(spell)

      spell |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |         12       30.00       30.00
          1 |         16       40.00       70.00
          2 |         12       30.00      100.00
------------+-----------------------------------
      Total |         40      100.00

. l , sepby(spell)

     +------------------------------------------------------+
     |      t      sales   spell   spell1   spell2   spell3 |
     |------------------------------------------------------|
  1. | 1990q1        100       0        1        0        0 |
  2. | 1990q2   97.84603       0        1        0        0 |
  3. | 1990q3   98.84029       0        1        0        0 |
  4. | 1990q4   100.8275       0        1        0        0 |
  5. | 1991q1   98.90981       0        1        0        0 |
  6. | 1991q2   100.9992       0        1        0        0 |
  7. | 1991q3   101.9653       0        1        0        0 |
  8. | 1991q4   104.1229       0        1        0        0 |
  9. | 1992q1   99.74297       0        1        0        0 |
 10. | 1992q2   102.3116       0        1        0        0 |
 11. | 1992q3   104.2382       0        1        0        0 |
 12. | 1992q4   105.8506       0        1        0        0 |
     |------------------------------------------------------|
 13. | 1993q1   102.8379       1        0        1        0 |
 14. | 1993q2    104.604       1        0        1        0 |
 15. | 1993q3   106.5201       1        0        1        0 |
 16. | 1993q4   106.9322       1        0        1        0 |
 17. | 1994q1   101.7508       1        0        1        0 |
 18. | 1994q2    107.551       1        0        1        0 |
 19. | 1994q3   106.7199       1        0        1        0 |
 20. | 1994q4   108.2395       1        0        1        0 |
 21. | 1995q1    103.035       1        0        1        0 |
 22. | 1995q2   107.8659       1        0        1        0 |
 23. | 1995q3   107.6514       1        0        1        0 |
 24. | 1995q4   107.3684       1        0        1        0 |
 25. | 1996q1   102.7467       1        0        1        0 |
 26. | 1996q2   107.6782       1        0        1        0 |
 27. | 1996q3   111.0517       1        0        1        0 |
 28. | 1996q4   112.0229       1        0        1        0 |
     |------------------------------------------------------|
 29. | 1997q1   103.7079       2        0        0        1 |
 30. | 1997q2   104.9966       2        0        0        1 |
 31. | 1997q3   110.0057       2        0        0        1 |
 32. | 1997q4   109.8376       2        0        0        1 |
 33. | 1998q1   104.5891       2        0        0        1 |
 34. | 1998q2   107.5621       2        0        0        1 |
 35. | 1998q3   109.6842       2        0        0        1 |
 36. | 1998q4   111.4588       2        0        0        1 |
 37. | 1999q1   108.9672       2        0        0        1 |
 38. | 1999q2   108.7656       2        0        0        1 |
 39. | 1999q3   112.9617       2        0        0        1 |
 40. | 1999q4   111.9473       2        0        0        1 |
     +------------------------------------------------------+

or just use factor variable notation on the spell variable.

[Ariel Karlinsky]

Sorry to hijack the thread, but I have a related (yet silly) question.
Assume my data has a variable which is stored as a string with values such as: "1998q4", "2003q2", etc.
How can I transform it to a proper quarterly-time variable for stata? I tried using the date() function, but the generated variable is missing on all observations.

[Nick Cox]

Ariel: No; please start a new thread.