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: frailty models


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: frailty models
Date   Sat, 30 Jul 2011 12:10:11 +0200

On Sat, Jul 30, 2011 at 10:56 AM, Farzaneh Amanpour wrote:
> I have fitted gamma model to the survival data, and maximum likelihood
> converged to the solution and I could get the result. But when I added
> frailty component (gamma or inverse-Gaussian) to the model, I got this
> error:
> “Could not calculate numerical derivatives – missing values encountered”
> My dataset doesn’t have any missing values, and also when I fit other
> distributions like lognormal with frailty component, I don’t encounter
> this error.

You can try specifying the estimated coefficients from the gamma model
without frailty as starting values using the -from()- option and/or
try the -difficult- or -technique()- options, but there is no
guarantee that these work. The gamma model is already very flexible
and when you add a frailty term to it it can easily happen that the
model is so flexible that the data does no longer contain the enough
information to estimate it.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany


http://www.maartenbuis.nl
--------------------------

*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index