Re: st: flagging significant values in a variable

[Partho Sarkar <partho.ss+lists@gmail.com>]

True, I had overlooked the earlier solutions- because the question
appears on 2 separate threads .  I answered the one w/o any answers at
the time, w/o having seen the other thread. ( Btw this raises some
issues about duplicate threads, possibly unintendedly so,  which often
confuse! )

The code was just a sketch of an idea- I assumed (mistakenly perhaps)
that the user would realize the need to qualify the if loops in
practice (start off the loops with a foreach statement to loop through
all the observations) .  The tolerance given is also only an example!
All this based on what I think is a legitimate interpretation of the
original question!

Partho

On Sat, Mar 3, 2012 at 8:22 PM, Nick Cox <njcoxstata@gmail.com> wrote:
> This post overlooks earlier solutions posted yesterday. I see no need
> to complicate anything by introduction of a tolerance, which seems
> based on an idea that the rates are exact decimals to 4 d.p.
>
> Also, the code won't work as intended because it confuses the -if-
> command and the -if- qualifier.
>
> FAQ     . . . . . . . . . . . . . . . . . . . . .  if command vs. if qualifier
>        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  J. Wernow
>        6/00    I have an if command in my program that only seems
>                to evaluate the first observation, what's going on?
>                http://www.stata.com/support/faqs/lang/ifqualifier.html
>
> Nick
>
> On Sat, Mar 3, 2012 at 2:10 PM, Partho Sarkar <partho.ss+lists@gmail.com> wrote:
>> Tim,
>>
>> I am afraid you haven't spelt it out very clearly! Based on one
>> possible interpretation, this would be one way to do it (shown only
>> for the LCI (renamed lci) variable):
>>
>> ---------------------------START CODE-------------------------------------------
>>
>> egen natlci=total(lci*(region==99)) // generates a value for each
>> obs., equal to national value)
>> local tol .0001  // define tolerance for "significantly lower or higher"
>> gen byte tag= .
>> if lci-natlci>`tol' {
>> replace tag=1
>> }
>> else if lci-natlci< -`tol' {
>> replace tag= 2
>> }
>> else {
>> replace tag = 0
>> }
>>
>> ---------------------------END CODE-------------------------------------------
>>
>> Hope this helps
>>
>> Partho
>>
>>                        From      Tim Evans <Tim.Evans@wmciu.nhs.uk>
>>                        To        "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
>>                        Subject   st: flagging significant values in a variable
>>                        Date      Fri, 2 Mar 2012 09:24:46 +0000
>>
>>                         Hi,
>>
>>                        I have a dataset that has variables of rates, LCI and UCI for a
>> number of regions in addition to a national average (rate, LCI, UCI)
>> so that it looks like this:
>>
>>                        rate            LCI             UCI             region
>>                        0.9727  0.9583  0.9849  1
>>                        0.9713  0.9523  0.9867  2
>>                        0.9835  0.9667  0.9971  3
>>                        0.9790  0.9741  0.9836  99
>>
>>                        What I would like to do is generate a flag beside each row that
>> will flag up entries where they are significantly higher (1) or lower
>> (2) or not significantly different (0) to region 99 - I'm unsure as to
>> the code here and would appreciate any advice. I'm using Stata 11.2.
>
> *
> *   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/