Re: st: Discrete time hazard model - Interval width

[gideluca@unical.it]

Dear Steve,
Thank you for your help.
The stock-sample includes all 2004 graduates at risk of finding a job  
(more precisely searching for a job and not pursuing other studies)  
who have been interviewed one year (2005), three years (2007) and   
five years (2009) after their  graduation. So, we initially select  
only those who in the baseline interview 2004 are looking for a job.  
These individuals can remain in their unemployment state, find a  
permanent job or be lost to follow-up.

We do not know exactly the date of the first real job they find, we  
just know if at the time of each subsequent interview they have a  
stable job or not. In particular, the relevant questions to identify  
failures are: ?Are you working at this moment? and ?Is your job  
stable??. Graduates finding a stable job, say, in the first interview  
are our failures. So, they are not at risk anymore and they are  
discarded from the analysis. In addition, for people who had never  
gotten a "real job",  the  last observed interview serves to define  
our censoring points.

What?s wrong with ignoring in a discrete time hazard model the fact  
that interviews are not administered regularly over time?


Def. Quota Steve Samuels <sjsamuels@gmail.com>:

> -
> Giuliana-
>
> You can use intervals of unequal lengths for a discrete or, in your
> case, grouped, survival analysis. But exactly what information on job
> history was obtained at each interview?
>
> If you obtained  more specific information the date of the first real
> job,  use that information to define your time variable.  For people
> who had never gotten a "real job",  the  last observed interview will
> serve to define censoring points.
>
> Stephen Jenkins has some excellent resources about survival analysis.
> http://www.iser.essex.ac.uk/study/resources/module-ec968
> including his Lecture notes:
> http://www.iser.essex.ac.uk/files/teaching/stephenj/ec968/pdfs/ec968lnotesv6.pdf
>
> Steve
>
> On Tue, May 18, 2010 at 8:51 AM,  <gideluca@unical.it> wrote:
> > Dear Stata Users,
> >
> > We are assessing the transition from university to the first real job,
> > taking into account individual characteristics and the family background. We
> > use survey data including graduates from the summer sessions in the year
> > 2004 who were interviewed at one, three and five years from graduation.  In
> > order to carry out the analysis we estimate a discrete time hazard model. We
> > have organised the data in such a way that for each student there are as
> > many data rows as there are time intervals over which the student is at risk
> > of finding the first job. Is it correct to use intervals of different width
> > (0-1; 1-3; 3-5)  given that we do not have yearly interviews?
> >
> >
> >
> >        +-----------------------+
> >        id   studytim   dead
> >        -----------------------
> > 279.    182          2      0
> > 280.    182          2      1
> > 281.    183          3      0
> > 282.    183          3      0
> > 283.    183          3      1
> >        -----------------------
> > 284.    184          1      1
> > 285.    185          3      0
> > 286.    185          3      0
> > 287.    185          3      1
> > 288.    186          3      0
> >        -----------------------
> > 289.    186          3      0
> > 290.    186          3      0
> >        +-----------------------+
> >
> > Thank you very much.
> > Giuliana
> >
> > ----------------------------------------------------------------
> > This message was sent using IMP, the Internet Messaging Program.
> >
> >
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/help.cgi?search
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
>
>
>
> --
> Steven Samuels
> sjsamuels@gmail.com
> 18 Cantine's Island
> Saugerties NY 12477
> USA
> Voice: 845-246-0774
> Fax:    206-202-4783
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/



----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/