You have interval censoring. Search for that.
m.p.
Nick Cox wrote:

See [ST] discrete and the works of Stephen Jenkins.
That said, it's not clear that the discreteness is
that crucial here.

Nick
n.j.cox@durham.ac.uk

Shon Hiatt

BACKGROUND

I have data on 1000 small firms from a survey that was administered in

1990. The survey had 140 questions answered by the entrepreneurs

themselves.

That same interview was administrered exactly a year later in 1991 to

only those firms who survived/continued from the original 1000

(N=600).

The interview was again administered three years later in 1994 to

firms that continued to exist from the original 1000 (N=350).

I don't have information of when the firms disappeared/died; I only

know that at some point between the administration of the 1st and 2nd,

and the 2nd and 3rd surveys that some disappeared. There is a firm

identification number to track the firms that survived. I also don't

have data on firms that disappeared--only on those that survived.

I am using the variables to explain survival.

QUESTION:

What is the best way to go about a survival analysis like this where
Time=1, 2, 4 and I don't know eactly when the firms disappeared

between the time intervals?

*

* 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/