Re: st: stock sample duration model with no follow-up

[Antoine Terracol <Antoine.Terracol@univ-paris1.fr>]

Margherita,

I might have misunderstood your problem, but here are my two cents:

If you only have one observation per individual, I don't think you can 
estimate any duration model.

If the ongoing duration if t at the time of the survey, and you observe 
no exit and have no follow-up for any individual , the likelihood for 
each spell will be one, because it is the probability that the duration 
is greater than t, conditionnal on the duration being greater that t : 
Pr[T>t|T>t]=1

The conditionning part is the stock sampling correction.


If your survey contains a retrospective calendar, you might be able to 
construct complete spells and get a sample on which you can estimate 
something.

Best,
Antoine



margherita.comola@upf.edu wrote:

> Dear statalister, 
> I have data on unemployment spells for all individuals who happen to be unemployed at the time of the survey. Thus, all observations are censored and I have no follow up. 
> how can I estimate this duration model in STATA?
>
> thanks alot
> Margherita
>
>
> *
> *   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/

--
Ce message a ete verifie par MailScanner
pour des virus ou des polluriels et rien de
suspect n'a ete trouve.

*
*   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/