1) For snapspan, I have:
-id: code of person identification
-fecha: date when person is observed
-ocu200: failure variable which states employment
condition:
ocu200==1 when person is employed
ocu200==2 when person is unemployed
ocu200==4 when person is out of labour force.
-fechaoc1: date when person began to work (escape of
unemployment)
-fechades: date when person began to search a job.
I put:
snapspan id fecha ocu200 fechades fechaoc1, gen
(fecha0)
The results appear like this:
Id Time 0 time1
1 . 10feb2004
1 10feb2004 10may2004
But by this way, later when I will stset my data,
the first record of each person would be lost and I
would have less observations. How can I snapspan my
data?
2) Then, for stset the data, I have:
- id: code of person identification
- time0: beginning of the span
- time1: ending of the span
- ocu200: failure variable (occurring at time1) which
states employment condition:
ocu200==1 when person is employed
ocu200==2 when person is unemployed
ocu200==4 when person is out of labour force
-fechaoc1= date when person began to work (escape of
unemployment)
-fechades =date when person began to search a job.
-X = enduring variables (sex, age, etc)
Additionally, people begin at risk when they are
unemployed and they exit analysis when they become
employed or when they get out labour force.
How can I stset my model considering that people can
have multiple failures and that usually, they begin
risk before they are observed?
Thanks,
Patricia
______________________________________________
Renovamos el Correo Yahoo!
Nuevos servicios, m�s seguridad
http://correo.yahoo.es
*
* 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/
