Re: st:survival data format

[Simon <scmoore.lists@googlemail.com>]

Getachew,

I think you might want to look at -snapspan- to start
getting your data into the right format:

input id time emp
 1 1 1
 1 2 1
 1 3 0
 1 4 0
 1 5 1
 2 1 1
 2 2 0
 2 3 1
 2 4 1
end

save temp.dta, replace
snapspan id time emp, generate(time0)

My version of snapspan insists that I save the data first.
Simon



On 22/03/2010 16:24, getachew Emrie wrote:
> I was trying to set  my data in to survival data format. My Interest
> is to look the duration of unemployment from CPS data.
> The data i have has the following form.
>
>    id   time     employment
>    1      1               1
>    1      2               1
>    1      3               0
>    1      4               0
>    1      5               1
>    2      1               1
>    2      2               0
>    2      3               1
>    2      4               1
>
> (employment=1 if employed and =0 if unemployed). My interest is to
> follow the person when he/she enters in to unemployment and look the
> time he/ she is going back to employment. Here failure is being
> employed (=1). when i try to set it up using stset command with id and
> failure=1,data starts with failure and stata ignores all the
> observations after first failure. it tells me that most of the data
> starts with failure. Is there any way tell stata to ignore failures
> before risk of being unemployed  ( for example in the above data  i
> want to ignore failures at time 1 and 2 and start at time 3 and then
> fail at time 5 for id=1).
>
> Thank you
> Getachew.
> *
> *   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/
*
*   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/