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

Re: st: fractional polynomial with glm and the cluster option


From   [email protected] (Roberto G. Gutierrez, StataCorp)
To   [email protected]
Subject   Re: st: fractional polynomial with glm and the cluster option
Date   Fri, 09 Jul 2004 09:10:57 -0500

Garry Anderson <[email protected]> asks:

> mfp is multivariable fractional polynomial.

> I am wishing to do mfp glm y x , fam(nb #) link(log) cl(cluster)

> The message -option cl() not allowed- appears

[...]

You have encountered a bug in -mfp-.  A subprogram of -mfp- runs under version
6, and thus any regression command (such as -glm-) called by this subprogram
is run under version 6 as well.  Most of the time this is not a problem since
estimation commands are modernized without regards to version control (the
modern version is usually the one you want anyway).  In the case of -glm-,
however, running under version 6 will give the much different version 6
release of -glm-, which did not allow the -cluster()- option.

-mfp- will be fixed so as to run its regression subcommand under the same
version as that which called -mfp- in the first place.  The fix will be made
available in a future ado update.

In the meantime, you can obtain what you want in two steps.

   1. run -mfp glm- without the cluster option and let it determine the 
      optimal fractional polynomial terms for you.

   2. run a final -glm- model with the terms obtained from 1., this time 
      including the cluster() option.  You can determine the terms to 
      be included in your model by examining the coefficient table from 1.

--Bobby
[email protected]
*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index