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: How to agreggate some specific observation to replace in another observation
From
Nick Cox <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: How to agreggate some specific observation to replace in another observation
Date
Wed, 21 Aug 2013 09:27:43 +0100
Sorry, but this strikes me as a bad idea for Stata. You want to
sprinkle your variable with subtotals. The effect would just be to
introduce double counting and to render the variable useless for
virtually any other analysis unless you excluded those observations
from all subsequent commands.
Assuming that there are no other variables, what I suggest rather is
to delete observations designated for subtotals. I guess -NACE- is a
string variable in which case
drop if strpos(NACE, "_")
destring NACE, replace
Then you are interested in looking at the distribution of -NACE- for
which there are many possible tabulation and graph commands.
Nick
[email protected]
On 21 August 2013 08:51, leila Peltier <[email protected]> wrote:
> Dear Stata users,
>
> As I entitled my message I want to replace one observation with the
> SUM of some other specific obseration in the best and efficient way.
> My original data are :
> VAR1 NACE
> 16_18
> 16
> 281.1 17
> 112 18
> 19
> 423.4 20
> .1 21
> 22_23
> 1713.9 22
> 702.1 23
> 24_25
> 2632.3 24
> 814.5 25
> 246.5 26
> 204.8 27
> 889.2 28
> 29_30
> 109.7 29
> 1.6 30
> .... and so on
>
> And I want to have :
>
> VAR1 NACE
> 393.1 16_18
> 16
> 281.1 17
> 112 18
> 19
> 423.4 20
> .1 21
> 2416 22_23
> 1713.9 22
> 702.1 23
> 3446.8 24_25
> 2632.3 24
> 814.5 25
> 246.5 26
> 204.8 27
> 889.2 28
> 111.3 29_30
> 109.7 29
> 1.6 30
>
> so for example 393.1 (NACE=16_18) is the sum of VAR1 if
> NACE==16&NACE==17&NACE==18
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/