Metcalfe, Paul
> I need to generate a time2 variable in my panel dataset
> which corresponds to
> quarterly time periods around an "event" which occurs in
> different months
> for each ID. My data looks like the following except for
> Time2 which I am
> unable to generate. The panel is unbalanced so it doesn't
> have observations
> for every month for every ID.
>
> e.g.
> ID Time Event Time2
> 1 1 . -1
> 1 2 . -1
> 1 3 . 0
> 1 4 1 0
> 1 5 . 0
> 1 6 . 1
> 1 7 . 1
> 1 8 . 1
> 1 9 . 2
> 1 10 . 2
>
> I've tried using explicit subscripting to do this using the
> following syntax
> but Stata cascades the changes all the way rather than changing one
> observation like I want it to.
>
> sort id time
> by id: gen time2=0 if event!=.
> by id: replace time2=0 if (time2[_n-1]==0)|(time2[_n+1]==0)
> by id: replace time2=-1 if
> (time2==.)*((time2[_n+1]==0)|(time2[_n+2]==0)|(time2[_n+3]==0))
> by id: replace time2=1 if
> (time2==.)*((time2[_n-1]==0)|(time2[_n-2]==0)|(time2[_n-3]==0))
>
> etc...
>
> Please can someone let me know how I can do this? Thanks in
> advance for any
> help anyone can give me.
Spread the event to other times for same id:
. bysort id : egen time2 = max(event * time)
Various other -egen- functions would also
do the job, on the understanding that the event
occurs just once for each id.
Centre on -time2- and scale to quarters:
. replace time2 = round(time - time2,3) / 3
Nick
[email protected]
*
* 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/
