On Tue, 2 Sep 2003 06:37:39 -0700 (PDT) Davood Souri <[email protected]>
wrote:
> Yes, I don't have "Failure" in my data set.
> Lancestar(1990) has called this type of data as stock
> samples. How can I use Stata to estimate duration
> models with stock samples?
It's not just the stock sampling aspect per se. Lancaster
(Econometrica, 1979) worked with stock sample data, but with interviews
at a later date, by which time some people had left unemployment. (This
is an example of 'stock sample with follow-up' otherwise known as
'delayed entry' or 'left truncation'.)
If there is stock sampling and no follow-up, then modelling is a very
complex business (you have to model differences in entry rates too) --
have a look at the likelihood that Nickell (Econometrica, 1979) writes
down to model his data.
But Nickell's data did include some spells that were completed (this
could occur because of the nature of his data, drawn from a national
household survey, rather than a sample from benefit administration
records).
You appear to have stock sampling and no failures. I don't see how you
can estimate duration models with this information alone.
Stephen
----------------------
Professor Stephen P. Jenkins <[email protected]>
Institute for Social and Economic Research (ISER)
University of Essex, Colchester, CO4 3SQ, UK
Tel: +44 (0)1206 873374. Fax: +44 (0)1206 873151.
http://www.iser.essex.ac.uk
*
* 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/
