Statalist The Stata Listserver


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

st: RE: ivtobit positive likelihood


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: ivtobit positive likelihood
Date   Sun, 4 Mar 2007 16:37:22 -0000

Why not? This is unusual, but possible. The underlying key here is
that a probability density for a continuous variable is _not_ itself
a probability. It is thus not bounded by 0 and 1, and will have 
units that are the reciprocal of those of your (unstated) response 
variable. Follow that through, and likelihood can exceed 1; 
and so log likelihood can exceed 0. So, on this evidence, there
is no problem to fix.

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

Richard Boylan
 
> I am using ivtobit for the first time and I am obtaining positive log
> likelihood. I did not think that this was possible. Any explanations
> and possible solutions? Thanks.

*
*   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