Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# RE: st: Panel Data Manipulation, divide all by a group of data??

 From "Martin Weiss" To statalist@hsphsun2.harvard.edu Subject RE: st: Panel Data Manipulation, divide all by a group of data?? Date Sat, 10 Jul 2010 19:59:27 +0200

```<>
-su, mean- is a good alternative if you want -quietly- and -r(mean)-

HTH
Martin

-original message-
Subject: Re: st: Panel Data Manipulation, divide all by a group of data??
From: Duha Altindag <daltin2@tigers.lsu.edu>
Date: 09-07-2010 23:42

I suppose your countries are also associated some time variable, since
this is a panel. Then you can do the following:

forval c=1/2 {
g col`c'_aboveUSA=.
forval t=1/3 {
qui su column`c' if country=="USA" & time==`t'
replace col`c'_aboveUSA= column`c'>=r(mean) & time==`t'
}
}

I guess this will work if you have a balanced panel, though.

On Fri, Jul 9, 2010 at 3:58 PM, Halit Akturk <halitakturk@gmail.com> wrote:
> Hi,
> I am a Stata 11 user. I am trying to do some data manipulation with a
> panel data (long form). Can someone advice me on how to do what I am
> trying to do in a more efficient way? I have a panel data of the
> following sort:
>           column1         column2
> CAN     10                     1
> CAN     55                     2
> CAN     12                     3
> UK       11                     4
> UK       12                     5
> UK       9                       6
> USA     16                     7
> USA     22                     8
> USA     25                     9
>
> What I want to is to divide all values in column 1 and column 2 for
> all countries by only the USA values in column 1!
> I did this computation in sort of a manual way by basically getting
> rid off all wariables other than USA column 1 values and saved that as
> a .dta file (say try.dta). Then I merged try.dta with the existing
> .dta file I have (above). This produced me a new column (say column 3)
> which contains USA column 1 values aligned with CAN and UK and USA
> column 1 values. Then I was able to compute by simply manipulating
> columns. My method is sort of manual and this requires me to have a
> seperate .dta file which I really don't like.
> Anyhow, I hope I am clear in explaining what I like to do. I want to
> learn whether there is an easier way of doing what I am trying to do
> in a more efficient and nicer way.
>
> Thank you all for your help.
>
> Halit Akturk
> Department of Economics
> Vanderbilt University
> *
> *   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/
>

*
*   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/

*
*   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/
```