Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: RE: RE: Replacing the missing observations
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: RE: RE: Replacing the missing observations
Date
Thu, 17 Nov 2011 23:29:33 +0000
Sorry, but I don't understand what your problem now is or in what
sense the results you show are wrong for the times you mention.
-tsset- is certainly going to make no difference to the effects of the
code I suggested as it makes no use of time series operators. Also,
your query refers to calculations _for the same time_, which are
nothing to do with time series operators.
Nick
On Thu, Nov 17, 2011 at 11:06 PM, Gokhan Yilmaz <[email protected]> wrote:
> Thanks Dr. Cox,
>
> I ran the command as you suggested and read the FAQ. However, I could not
> figure out why it worked for only some of the time values. It replaced a
> zero for the ewdiffret instead of the actual values in 2007m6 through
> 2007m12 and 2008m2. I had tsset the data but I also tried it after "tsset,
> clear" and the same thing happened. For the other time values it worked
> perfectly. Also it may seem like I am trying to replace for only one
> observation in a given month but indeed there are around 345 observations
> for each month to be replace with the same ewdiffret value in that given
> month.
>
>
> group time ewportret ewdiffret
> 0 2007m4 .041679
> 1 2007m4 .0375245 -.0041545
> 0 2007m5 .0449738
> 1 2007m5 .0397919 -.0051819
> 0 2007m6 -.0240277
> 1 2007m6 -.0188678 .0051599
> 0 2007m7 -.064496
> 1 2007m7 -.0456775 .0188185
> 0 2007m8 .0023347
> 1 2007m8 .0104131 .0080783
> 0 2007m9 .0237398
> 1 2007m9 .0209782 -.0027617
> 0 2007m10 -.0086541
> 1 2007m10 .0190929 .0277471
> 0 2007m11 -.0589931
> 1 2007m11 -.0399782 .0190149
> 0 2007m12 -.0302353
> 1 2007m12 -.0077903 .022445
> 0 2008m1 -.0117908
> 1 2008m1 -.0418619 -.0300711
> 0 2008m2 -.0362031
> 1 2008m2 -.0311441 .0050591
> 0 2008m3 .0164306
> 1 2008m3 -.0015971 -.0180278
>
> Regards,
>
> Gokhan
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Nick Cox
> Sent: Wednesday, November 16, 2011 5:24 AM
> To: '[email protected]'
> Subject: st: RE: Replacing the missing observations
>
> This is an FAQ.
>
> FAQ . . . . . . . . . . . . . . . . . . . . . . . Replacing missing
> values
> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . N. J.
> Cox
> 9/05 How can I replace missing values with previous or
> following nonmissing values or within sequences?
> http://www.stata.com/support/faqs/data/missing.html
>
> In this example
>
> bysort time (ewdiffret) : replace ewdiffret = ewdiffret[1]
>
> Note that -replace- is a command (not a function).
>
> would work.
>
> Nick
> [email protected]
>
> [email protected]
>
> group time ewportret ewdiffret
>
> 0 2007m4 .041679 .
> 1 2007m4 .0375245 -.0041545
> 0 2007m5 .0449738 .
> 1 2007m5 .0397919 -.0051819
> 0 2007m6 -.0240277 .
> 1 2007m6 -.0188678 .0051599
>
> The variable ewdiffret is the difference in ewportret between group 1 and
> group 0 for each time variable (thanks to Dr. Cox for the help with this).
> My question is how can i now replace all empty cells under the ewdiffret in
> the same month with the single ewdiffret value in that given month. Indeed i
> have too many empty cells in a given month and i want to replace them all
> with the single ewdiffret i have for that month.
>
> For example: I would like to have the first empty cell under ewdiffret equal
> to -.0041545 as the time variable for both is 2007m4 and the second empty
> cell equal to -.0051819 as their time variable is 2007m5. Using too many
> replace functions takes a long time and may cause transcription errors.
>
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/