Statalist


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

RE: st: GLLAMM


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   stata list <statalist@hsphsun2.harvard.edu>
Subject   RE: st: GLLAMM
Date   Tue, 16 Feb 2010 17:55:43 +0000 (GMT)

--- Alberto R. Osella wrote:
> I'm trying to fit a GLLAMM to data <snip> but there are 
> problems with the convergence.

Could be many things. One thing that struck me was the 
line: -gen lvita = log(vita+0.001)- in your code. This 
can very easily turn your zeros in big outliers, which 
can cause trouble with convergence. For example in the
example code below you can see that the zeros are put
way outside the normal range of the data.

*----------- begin example -------------
sysuse nlsw88, clear
gen ln_tenure = ln(tenure + .001)
spikeplot ln_tenure
*------------ end example --------------

Hope this helps,
Maarten

PS. formatted text (rich text, xml, html) don't format
well on the list. Your message looks for example like: <http://www.hsph.harvard.edu/cgi-bin/lwgate/STATALIST/
archives/statalist.1002/date/article-721.html> For that
reason the Statalist FAQ askes everyone to post their
messages as plain text. See number of 4 of:
<http://www.stata.com/support/faqs/res/statalist.html#toask>

--------------------------
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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index