Bookmark and Share

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: Problems with execution of conditional branching..


From   Nick Cox <[email protected]>
To   [email protected]
Subject   Re: st: RE: Problems with execution of conditional branching..
Date   Thu, 14 Mar 2013 11:58:48 +0000

A footnote to Sarah's very clear explanation:

The help does warn that the -if- qualifier and the -if- command are different.

See also

FAQ     . . . . . . . . . . . . . . . . . . . . if command versus if qualifier
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  J. Wernow
        4/05    I have an if or while command in my program that
                only seems to evaluate the first observation.
                What's going on?
                http://www.stata.com/support/faqs/programming/
                if-command-versus-if-qualifier/

That said, the FAQ has it more or less backwards in terms of what
people ask. The frequently asked question is much more usually
something like "I don't understand what the -if- command is doing, as
it is not giving the results I expect" and the answer is then often
"the -if- command is evaluating your first observation only -- which
is unlikely to be what you want -- and so you need the -if- qualifier
instead".

Regardless, this difference is so often misunderstood that the help
could usefully be strengthened on this point, if only by referring
explicitly to the FAQ.

My understanding is that SAS differs on this point and my experience
is that SAS users coming to Stata are often bitten by the difference.

Nick

On Thu, Mar 14, 2013 at 1:22 AM, Sarah Edgington <[email protected]> wrote:

> The problem is that the if statements here are only evaluating your first
> observation.  See -help ifcmd- and -help if- for more details.
> The thing that I think isn't entirely clear from the helpfile is that the
> programming if checks the conditional statement once and if it's true it
> takes the action in the braces.  Unless otherwise specified it evaluates the
> first observation in the data when the conditional involves a variable.
> To get an observation by observation evaluation of a conditional you have to
> use the if expression after a command.
> So -gen event_month=substr(opendate,6,1) if comma_3==2 & comma_2==1- will
> check that condition for each observation and create the relevant variable.
>
> Contrast that with a simplification of your first conditional (written as a
> single line instead of in braces)
> -if comma_3==2 & comma_2==1 gen event_month = substr(opendate, 6,1)-
> Here Stata is only evaluating the conditional once.  Since it involves
> variables (which, by virtue of being variables are going to be different
> across observations) there has to be some rule about what observation is
> used to evaluate this statement.  Since you haven't specified an observation
> number to evaluate, it defaults to the first observation.
>
> So running your code will perform only one of the four possible sets of
> commands you specified.  Which one it performs depends entirely on the first
> observation in the data.
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index