Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: negative binomial regression accepts non-integer continuous outcomes?


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: negative binomial regression accepts non-integer continuous outcomes?
Date   Fri, 31 Aug 2012 01:47:49 +0100

The short answer to Jason's question is neither. I am puzzled that

1. Jason thinks that Stata would change your data, but not make that clear.

2. Jason didn't experiment with truncating or rounding his data to
check out his fears.

3. No-one in practice seems puzzled that Stata doesn't object to
integers when applying tests or fitting models that are usually
explained as based on normality assumptions.

Nick

On Fri, Aug 31, 2012 at 1:02 AM, Owen Gallupe <ogallupe@gmail.com> wrote:
> Hi Jason,
>
> I ran up against this same issue a few months back. The discussion
> thread at the link below should be informative.
>
> http://www.stata.com/statalist/archive/2011-12/msg00182.html
>
> Owen
>
>
> On Thu, Aug 30, 2012 at 7:16 PM, Jason Bond <jbond@arg.org> wrote:
>> I'm running negative binomial regression using nbreg and it seems to accept dependent variables that are positive and continuous (i.e., they are not required to be positive integers).  Is Stata truncating or rounding these dependent variable values to integer values before estimating the model?  Thanks for any guidance,

*
*   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/


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