Or alternatively something like
bysort ind (t): gen part2 = 2 * part[2] + part[1] + 1
But do note the (t) in
bysort ind (t):
Otherwise there's a major risk of inconsistent time
order.
Nick
[email protected]
R.E. De Hoyos
>
> Gang,
>
> I am sure there are more elegant ways to do it, but this should work:
>
> gen part2=.
> bys ind: replace part2=1 if part==0 & part[_n-1]==0
> bys ind: replace part2=2 if part==1 & part[_n-1]==0
> bys ind: replace part2=3 if part==0 & part[_n-1]==1
> bys ind: replace part2=4 if part==1 & part[_n-1]==1
>
> Each individual will have only one value (1 to 4) in the
> final year (t+1)
> and a missing value in the innitial year (t). If you want to
> replace the
> missing values with the ones generated by part2 then you simply:
>
> bys ind: egen part_f=mean(part2)
>
> Notice that if have more time periods, then you would need a
> better coding
> to account for all possible combinations.
> From: "is300gp" <[email protected]>
> To: <[email protected]>
> Sent: Wednesday, July 28, 2004 7:35 PM
> Subject: st: panel data command
>
>
> > Hi, dear all:
> >
> > I have simple question about data manipulation is panel
> data: the data is
> > in standard panel stata format, with i = individual and t=2. the
> > participation is dummy on work participation in either
> time; so we have
> four
> > categories of individual on work participation: those
> participate in both
> > period, neither period, only in the first period and only
> in the second
> > period. So my questions is how can I take out the
> individuals falling into
> > these four categoris?
> >
> > I used to export into excel and use macro to do the job,
> but with so many
> > time of export and inport, I'm kind of running out of gas.
> Could anybody
> > give a better suggestion?
*
* 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/
