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: Offset variable


From   Nick Cox <njcoxstata@gmail.com>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Offset variable
Date   Mon, 9 Dec 2013 11:07:53 +0000

One of the things you learn as a journal editor is that authors very
rarely want to write the papers you want them to write. They have the
awkward habit of preferring their own ideas. I don't think anyone
wrote that paper. As the best ideas in statistics don't really go out
of date, it would be as welcome in 2013 or 2014 as in 2004.

Nick
njcoxstata@gmail.com


On 9 December 2013 05:43, Richard Williams
<richardwilliams.ndu@gmail.com> wrote:
> At 10:53 PM 12/8/2013, Stephen A. Crowder wrote:
>>
>> I am developing an ordinal logistic regression model examining a factor
>> with several levels on my dependent variable. That part is straight forward
>> enough but I have a main independent variable that ide like to control for
>> in the examination of all of the other independent variables. My question is
>> would I need to make this an "offset variable" in my models? I've been
>> looking through the literature and there's no clear answer available.
>>
>> Sent from my iPhone
>
>
> First off, if you have a new topic you should start a new thread rather than
> reply to an old post. Even if you change the topic the threading gets thrown
> off in the archives. Also the totally unrelated extraneous text from the old
> thread that you quoted is distracting.
>
> I think you probably just want to include the variable as another
> independent variable. In Poisson and related models offset is sometimes used
> as an alternative to the exposure option, e.g. it provides a way of dealing
> with cases that have been at risk for differing amounts of time. I don't
> think I have ever seen an example that used offset in an ologit model. I'd
> be curious to know if there are situations in which it makes sense.
>
> I don't know if the offer is still good, but 9 years ago Nick Cox said SJ
> would be interested in a paper that discussed the practical uses of offset
> with commands like ologit:
>
> <http://www.stata.com/statalist/archive/2004-10/msg00002.html>http://www.stata.com/statalist/archive/2004-10/msg00002.html
>
> If somebody did successfully take Nick up on his offer I'd like to know what
> the paper was.
*
*   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