Replacing a variable which has an initial value

Question

I have a panel data set (the identifiers are the variables of cc and year).

I have all the values for the Y variable but I have only the initial (X_{1996}) values of the X variable.

I would like to replace the missing values of X with this pattern:

X_{1997} = Y_{1996} + 0.95*X_{1996} 
X_{1998} = Y_{1997} + 0.95*X_{1997}
.
.
.
X_{2015} = Y_{2014} + 0.95*X_{2014}

I would like to iteratively replace X starting from 1997, that is using the last generated value.

Is it possible to write a loop for this purpose?

Example data:

clear 
input str3 cc   year     X         Y
"ABW" 1996 467.5953  28.46004
"ABW" 1997        .  30.23142
"ABW" 1998        .  31.70577
"ABW" 1999        . 30.870903
"ABW" 2000        . 25.495903
"ABW" 2001        .  23.71254
"ABW" 2002        . 26.906475
"ABW" 2003        . 29.215036
"ABW" 2004        .  28.77914
"ABW" 2005        . 33.581017
"ABW" 2006        .  35.16129
"ABW" 2007        .  32.51442
"ABW" 2008        . 33.251934
"ABW" 2009        .  30.29287
"ABW" 2010        . 28.931993
"ABW" 2011        .   28.5057
"ABW" 2012        . 26.647564
"ABW" 2013        . 23.999134
"ABW" 2014        . 22.917984
"ABW" 2015        . 21.481943
"AFG" 1996  362.353 11.51442
"AFG" 1997        . 11.557285
"AFG" 1998        . 12.19376
"AFG" 1999        . 14.557285
"AFG" 2000        . 12.906254
"AFG" 2001        . 13.156855
"AFG" 2002        . 11.557285
"AFG" 2003        . 15.758896
"AFG" 2004        . 19.005346
"AFG" 2005        . 23.193766
"AFG" 2006        . 26.407976
"AFG" 2007        . 20.068926
"AFG" 2008        . 19.021326
"AFG" 2009        . 17.991436
"AFG" 2010        . 17.959072
"AFG" 2011        . 16.674591
"AFG" 2012        . 16.906254
"AFG" 2013        . 17.343513
"AFG" 2014        . 18.156855
"AFG" 2015        . 19.454687

For best results with panel data in Stata, xtset or tsset your data first and exploit time series operators, which not only respect panels but also do the right thing if there are gaps in your data. For that you need a numeric panel identifier, which encode can do for you. — Nick Cox
– Nick Cox, Commented Mar 18, 2019 at 11:57

score 3 · Accepted Answer · 2019-03-18 11:56:22Z

If you want to calculate the missing values of X by using the last generated observation, then you can do so as follows:

bysort cc (year): replace X = Y[_n-1] + 0.95 * X[_n-1] if _n > 1

list, sepby(cc) 

     +----------------------------------+
     |  cc   year          X          Y |
     |----------------------------------|
  1. | ABW   1996   467.5953   28.46004 |
  2. | ABW   1997   472.6756   30.23142 |
  3. | ABW   1998   479.2732   31.70577 |
  4. | ABW   1999   487.0153    30.8709 |
  5. | ABW   2000   493.5355    25.4959 |
  6. | ABW   2001   494.3546   23.71254 |
  7. | ABW   2002   493.3494   26.90648 |
  8. | ABW   2003   495.5884   29.21504 |
  9. | ABW   2004    500.024   28.77914 |
 10. | ABW   2005    503.802   33.58102 |
 11. | ABW   2006   512.1929   35.16129 |
 12. | ABW   2007   521.7445   32.51442 |
 13. | ABW   2008   528.1717   33.25193 |
 14. | ABW   2009    535.015   30.29287 |
 15. | ABW   2010   538.5571   28.93199 |
 16. | ABW   2011   540.5613    28.5057 |
 17. | ABW   2012   542.0389   26.64756 |
 18. | ABW   2013   541.5845   23.99913 |
 19. | ABW   2014   538.5045   22.91798 |
 20. | ABW   2015   534.4972   21.48194 |
     |----------------------------------|
 21. | AFG   1996    362.353   11.51442 |
 22. | AFG   1997   355.7498   11.55729 |
 23. | AFG   1998   349.5196   12.19376 |
 24. | AFG   1999   344.2373   14.55729 |
 25. | AFG   2000   341.5828   12.90625 |
 26. | AFG   2001   337.4099   13.15685 |
 27. | AFG   2002   333.6962   11.55729 |
 28. | AFG   2003   328.5687    15.7589 |
 29. | AFG   2004   327.8992   19.00535 |
 30. | AFG   2005   330.5096   23.19377 |
 31. | AFG   2006   337.1778   26.40798 |
 32. | AFG   2007   346.7269   20.06893 |
 33. | AFG   2008   349.4595   19.02133 |
 34. | AFG   2009   351.0078   17.99144 |
 35. | AFG   2010   351.4489   17.95907 |
 36. | AFG   2011   351.8355   16.67459 |
 37. | AFG   2012   350.9183   16.90625 |
 38. | AFG   2013   350.2787   17.34351 |
 39. | AFG   2014   350.1082   18.15685 |
 40. | AFG   2015   350.7597   19.45469 |
     +----------------------------------+

Collectives™ on Stack Overflow

Replacing a variable which has an initial value

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related