Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: pseudo R2s for Generalized Linear Models


From   n j cox <n.j.cox@durham.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: pseudo R2s for Generalized Linear Models
Date   Thu, 19 Jul 2007 17:55:01 +0100

Some or all of the material at

<http://www.stata.com/support/faqs/stat/rsquared.html>

may be of use or interest to you.

On the whole, I regard the brandishing of these
single figures of merit as an unsavoury macho exercise,
along the lines of "my model is really good
(considering)!" or (in some weak sciences)
"my model is pretty lousy (but not nearly
as lousy as it could be)!". It would be better
to devote more attention to model diagnostics,
especially examination of residuals, and
consideration of scientific or practical
interpretation.

That said, testosterone is in my system too,
and I have contributed -glmcorr-, which is
downloadable from SSC. This is not intended
as producing a large battery of measures, just
a couple that I find useful on occasion.

Of more importance than that program is the
paper that inspired it

Zheng, B. and A. Agresti. 2000. Summarizing the predictive power of a
generalized linear model. Statistics in Medicine 19: 1771-1781.

which I commend.

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

agostino@unical.it

I was wondering if there exists a routine to calculate (several) pseudo R2s for glm regressions. Currently I am calculating the pseudo R2 dividing the maximum value of the log-Likelihood function of the fitted model by the maximum log-likelihood of a model containing only a constant.
*
* 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