Statalist


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

RE: st: Perfect predictors of an occurrence (survival analysis)


From   "Luis Ortiz" <[email protected]>
To   <[email protected]>
Subject   RE: st: Perfect predictors of an occurrence (survival analysis)
Date   Thu, 13 Mar 2008 16:06:47 +0100

This is precisely what happens, Maarten. As you said, the final record
ALREADY records the type of contract for the next job.

In this sense, type of contract is not exactly like sex. Sex certainly
doesn't change during the risk period... but it doesn't change afterwards
either. Thus, it doesn't predict the outcome, whereas type of contract
predicts it perfectly... to my bad fortune.

Thanks again for thinking over it, Maarten

Luis

-----Mensaje original-----
De: [email protected]
[mailto:[email protected]] En nombre de Maarten buis
Enviado el: jueves, 13 de marzo de 2008 15:48
Para: [email protected]
Asunto: RE: st: Perfect predictors of an occurrence (survival analysis)

--- Luis Ortiz <[email protected]> wrote:
> > The problem is that some of my covariates, seemingly, do not change
> > their value during t. The type of contract does not change while
> > the individual holds a job for which s/he is over-educated.
> > 
> > In other words, the type of contract perfectly predicts my outcome
> > 
> > As you'll see next, whenever I run the model with sex and age
> > dummies, everything is OK. Yet, if I add to the model the 
> > type-of-contract dummies....

--- Maarten buis <[email protected]> wrote:
> I don't know the answer, but I think I can eliminate one potential
> cause, which I hope is sorta helpful. You claim the the prolem is
> that the contract doesn't change over time. This can't be  the
> problem, because you report that the model works OK when you add sex,
> and this presumably doesn't change over time either. 

Maybe the final record of each spell already records the type of
contract for the next job. I can imagine how a process of generating
spells would result in that, and how such a structure could lead to
perfect predictions (though I am not sure).

-- Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      __________________________________________________________
Sent from Yahoo! Mail.
The World's Favourite Email http://uk.docs.yahoo.com/nowyoucan.html
*
*   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/

*
*   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