Re: st: another aflogit issue

[n j cox <n.j.cox@durham.ac.uk>]

-aflogit- is a user-written program published in the STB:

STB-42  sbe21  . Adjusted pop. attributable fractions from logistic 
regression
        (help aflogit if installed)  . . . . . . . . . . . . . . . . 
A. Brady
        3/98    pp.8--12; STB Reprints Vol 7, pp.137--143
        estimation of the attributable fraction (AF) from within a
        logistic regression framework, thus enabling confounders to
        be taken into account, and allows estimation of the summary
        AF for a set of exposures

It does no harm and some good if this is made clear, as is indeed
advised in the FAQ.

I believe that the author, Tony Brady, is
not a member of Statalist, but he may be contactable as someone behind
http://www.sealedenvelope.com/

What you tell us boils down to this: you tried the command
with various variables, and it worked with some and not others.
Without knowing more about the data or exactly what you typed,
this really isn't much information to go on.

You could try -set trace on- to see where -aflogit- kicks
you out.

Nick
n.j.cox@durham.ac.uk


Marc Campo

---------------------------------- #2
I keep running LR and then typing in aflogit for
particular variables specifying a reference level.
For some variables, no problem.  For others, it says
variable not found or abbreviation is incorrect.  I am
typing in the right names.

the variable xfer runs.  the variables job_strain and
manual_total do not.  Anyone know why?
----------------------------------

---------------------------------- #1
> I am looking at a series of ordinal indep variables
> individually while adjusting for background factors
> with logistic regression.  I am entering these
> varaibles as continuous as opposed to dummy coding.
>
>
> I want to use AFLOGIT for attributable fractions.  I
> am not sure how this works with a continuous indep
> variable however.  My vars are coded from 1-4 or
> 1-5.
> 1 in each case is the lowest category.  The AF
> instructions state to make the reference cat your
> non
> exposure category, so I have used 1.  Is the result
> the AF for all other categories vs. 1 in that case?
>
> That's not what I want.  But with so few categories,
> I
> can't just take the middle value either. If it isn't
> just 1 vs. all others, then my results should be
> fine.
*
*   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/