Re: st: Creating a variable with the highest value from two othervariables

[Christer Thrane <lineea@broadpark.no>]

Thanks Joseph, but I can't get it to work. Can you se what is wrong in the 
assert-command

. sum X Y

   Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
          X |       364    6970.604    6315.697        100      45000
          Y |       360    3978.485    4844.969         .5      50000

. list X Y if X< Y

    +---------------+
    |     X       Y |
    |---------------|
30. |  2500    3000 |
34. |  7000       . |
37. |  1000    2000 |
54. |   700    1200 |
127. |   100     150 |
    |---------------|
137. | 20000   50000 |
162. |  2000       . |
209. |  2000    4000 |
254. |  1000       . |
276. |  2000    3000 |
    |---------------|
308. | 10000       . |
335. |   900    1090 |
441. | 30000       . |
    +---------------+

. assert !missing(X) & !missing(Y)
108 contradictions in 467 observations
assertion is false
r(9);

. generate byte Zstar = cond((X >= Y), X, Y)
(108 missing values generated)

. assert Zstar == Z

. sum Zstar

   Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
      Zstar |         0


Thanks again,

Christer








----- Original Message ----- 
From: "Joseph Coveney" <jcoveney@bigplanet.com>
To: "Statalist" <statalist@hsphsun2.harvard.edu>
Sent: Sunday, October 14, 2007 12:59 AM
Subject: Re: st: Creating a variable with the highest value from two other 
variables



> Christer Thrane wrote:
>
> I have two income measures, X and Y. I want to create a variable, Z, that 
> is
> equal to X for those obs where X>=Y and equal to Y for those obs where 
> Y>X.
> Like this:
>
> id    X    Y       Z
> 1    12    11    12
> 2    10    10    10
> 3    10    12    12
>
> Is there a simple way?
>
> --------------------------------------------------------------------------------
>
> You can use -cond()-.  You can nest -cond()-s to handle missing values.
>
> See also -egen-.
>
> Joseph Coveney
>
> clear *
> input byte id byte X byte Y byte Z
> 1    12    11    12
> 2    10    10    10
> 3    10    12    12
> end
> assert !missing(X) & !missing(Y)
> generate byte Zstar = cond((X >= Y), X, Y) // Here
> assert Zstar == Z
> exit
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/ 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/