Re: st: identify group of observations

[Doug Owens <dho@tamu.edu>]

At 08:54 PM 4/28/2004 -0500, you wrote:
> At 06:08 PM 4/28/2004 -0400, MITRA PINAKI  (MAR1PXM) wrote:
> > Hello List,
> >
> > I have a data set where I need to group observations. The identifying flag
> > is two character string which can be either "01" or "02". Each "01"
> > observation is followed by "02", so the flag for the first obs is "01" and
> > the very next one's is "02". But these are always not in pair; order can be
> > "01" and "02"; "01", "02", "02"; "01", "02", "02, "02" and so on. All the
> > "02" obs. followed by "01" belong to one group. Is there any way I can
> > generate an identifier that indicates these groups?
> Do you have that backwards? i.e. All the "01" obs. followed by "02" belong 
> to one group.  That would have been my guess based on your examples.
Assuming you want to create grp like:
Flag  grp
01      1
02      1
01      2
02      2
02      2
01      3
02      3
02      3
02      3

Then you can just
gen grp = flag=="01"
replace grp = sum(grp)















> *
> *   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/
*
*   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/