--- On Wed, 29/7/09, Lu, Zhenyan wrote:
> I have a question regarding the note warning after running
> discrete-time logit model.
> After the result table reported in Stata, sometimes there
> is note that says
> "Note: 61 failures and 0 successes completely determined."
> All others are regular.
>
> I checked relevant Stata guidance that basically it says
> this may be caused by empty cells. So I deleted all empty
> cells and rerun models, but still with this kind of
> problem.
Empty cells means empty cells in a cross-tabulation of
your dependent and independent variables not in your
data-matrix. It could for instance occur when in a year
there are no events.
For a more regular way of doing a discrete time survival
analysis see:
http://www.iser.essex.ac.uk/iser/teaching/module-ec968 .
I am actually not sure if this will solve your problem,
but at least you will be able to present your results in
a metric that is more common to survival analysis people
(hazard ratios rather than odds ratios), which may make
it easier to "sell" your paper.
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
