Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: tabulating frequency records


From   "Zekveld, Cornelia" <c.zekveld@imperial.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: tabulating frequency records
Date   Thu, 10 Jul 2003 17:49:47 +0100

Nick,

I'm aware that table doesn't allow non-integers.  So when I used 0 for those
freq <0.5 and tabulating them with -table-, this command didn't distinguish
between zero frequencies and missing combinations.  I need a command that
can distinguish between the two and found that tabdisp works fine, except I
cannot get totals - that's why I worked with table since this command gives
me totals, it's deciding which one to use since neither works perfectly.
I've decided to go with tabdisp and transporting the tables to another
package.

Thanks for you help
Cornelia



-----Original Message-----
From: Nick Cox [mailto:n.j.cox@durham.ac.uk]
Sent: 10 July 2003 17:25
To: statalist@hsphsun2.harvard.edu
Subject: st: RE: RE: RE: tabulating frequency records


Cornelia Zekweld 
 
> > --------------------------------------------------------------
> >                                          |       agecat
> >                                AgentName | 16-39  40-64    65+
> > -----------------------------------------+--------------------
> >                                A........ |     1      2      .
> >                                 C....... | 9,999      1      .
> >                                   C..... |     .      1  9,999
> >                            I............ |     .  9,999      .
> >                                   N..... |     1  9,999      .
> >                             O........... |     1      2  9,999
> > --------------------------------------------------------------
> >
> 
> I will have a look at the tabcount command - thanks.
> 
> The reason why the avg. freq. <0.5 is that it is annual 
> averages where e.g.,
> over a period of 5 years there are only 2 occurrences of 
> the disease and so
> dividing this by 5 gives you 0.4 per year - which then is 
> rounded to zero.
> 
> This causes another problem in that if I add col or row to 
> get totals by
> age-group, the zero's of course do not add anything to the 
> totals (where
> actually it was 0.4). Having say 3 such entries, the 
> overall total should
> increase by 1.2 (rounded to 1), but now I lose this.  
> 
> I have an immense number of tables to produce, otherwise I 
> could have worked
> out the annual average totals from the period-specific totals.

Sorry, I didn't make myself clear. I understand how 
averages < 0.5 arise. What I said was that I didn't know what 
such averages _entailed_ for -table-. In fact, I am not 
aware that they entail anything for -table-. 

-table- won't accept frequency weights which are not 
integers, so I still don't understand what you are doing
with -table-. Perhaps someone else does. 

Nick 
n.j.cox@durham.ac.uk 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index