Statalist


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

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


From   Antoine Terracol <Antoine.Terracol@univ-paris1.fr>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: stock sample duration model with no follow-up
Date   Thu, 20 Dec 2007 17:47:40 +0100

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/



© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index