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

Re: st: anscombone resid after glm negbin


From   rgutierrez@stata.com (Roberto G. Gutierrez, StataCorp.)
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: anscombone resid after glm negbin
Date   Tue, 11 Jun 2002 11:05:24 -0500

Back in early May, Mark Boye <mark.boye@pharmacia.com> asked:

>> Following Hardin & Hilbe, I have estimated a negative binomial with glm
>> -fam(nb, #) - in order to use GLM diagnostics.  Thereafer, again following
>> H&H, I attempted residual analysis with the use of Anscombe residuals.
>> Unfortunately, however, the predict command returns only missing values.
>> Why is this so?  Am I missing something?

and I responded:

> Mark seems to have uncovered an error in the calculation of Anscombe
> residuals after -glm- used with the negative binomial family.  We are
> currently working to confirm that this is indeed a bug (and our initial
> indications are that it most likely is), and on the fix.

> I'll report back to the list when I have something more definite to say on
> this matter.

This was a bug.  An incorrect formula was being used in the calculation of
Anscombe residuals (-predict, anscombe-) after -glm- with either a negative
binomial or a non-Bernoulli binomial family.

We have fixed this, and you can gain access to the fix by making your Stata
executable/ados up-to-date.

--Bobby
rgutierrez@stata.com
*
*   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