Re: st: generating variable indicating presence/nonpresence in dataset by observation criteria

[Nick Cox <njcoxstata@gmail.com>]

Look at -tsset- and then -tsfill-.

Nick


On 15 Jul 2011, at 09:54, Erik Aadland <erikaadland@hotmail.com> wrote:

> Dear statalist.
>
> I have an unbalanced panel dataset in which different actors  
> (agency_id) are present with a project (st_project_id) in the  
> dataset in some months (yearmonth) but not in other months. I would  
> like to estimate the likelihood that each actor (agency_id) is  
> present with a project (st_project_id) in yearmonth. Some actors  
> enter more projects, and are consequently more likely to be present  
> in yearmonths with projects.
>
> To do this, I am therefore trying to generate a variable for each  
> agency_id that takes on a value of 1 if the agency_id is present in  
> the dataset a given month, and 0 if the actor is not. Is this  
> possible? Or are there other ways for me to accomplish my objective?
>
> Here is an example dataset structure:
>
> yearmonth   agency_id   st_project_id
> 1           1           1
> 1           2           2
> 1           4           3
> 2           3           4
> 2           5           5
> 2           1           6
>
> In this example, agency_id 3 and 5 would get a value of 0 for  
> yearmonth 1 and a value of 1 for yearmonth 2. Agency_id 1 would get  
> a value of 1 for both yearmonths. Agency_id 2 would have a value of  
> 1 for yearmonth 1 and 0 for yearmonth 2, and so on.
>
> Any suggestions on this would be very much appreciated.
>
> Kind regards, Erik.
> *
> *   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/