Jesse--
Most of the relevant user-written contributions are collected in
-egenmore-, but almost any commands for this kind of work other than
-by-, -sort-, -gen, -replace-, and -egen- are just shortcuts
(including -gsort-) and in many cases it is more efficient to work
using only those 5 commands.
Your example is unclear--you seem to assert that the 7 obs belong to
the same surgeon, but the Surgeon variable differs--and your sort
order is not specified in full. Perhaps the "more" on the subject
line indicates that necessary info is in prior emails; I don't
remember them. Also, why use illegal names for variables? You are
imposing an extra burden on those who are supposed to help by not
providing a clean example of the beginning data and the target data.
input Surgeon Hosp Cases NumofHosp
1 1 4 1
1 1 4 1
1 1 4 1
1 1 4 1
1 1 2 2
1 1 2 2
1 2 1 2
2 1 4 1
2 1 4 1
2 1 4 1
2 1 4 1
2 2 2 2
2 2 2 2
2 2 1 2
end
g Patient=_n
g nc=-Cases
bys Sur (nc): g kp=(Hosp!=Hosp[_n-1] & _n>1)
by Sur: g keepers=sum(kp)
li, noo sepby(S)
On 5/16/07, Moore, Jesse <[email protected]> wrote:
I have data about surgeons who were operated on at different hospitals.
I'd like to keep the information from the hospital with the most cases:
Patient Surgeon Hosp Cases@hosp NumofHosp
A A 1 4 1
B A 1 4 1
C A 1 4 1
D A 1 4 1
E B 1 2 2
F B 1 2 2
G B 2 1 2
I've used gsort so Cases@hosp is in descending order. I'd like to have
a way of creating a new var (keepers) that would identify patients A-F
but not patient G (second hospital for the surgeon and the hospital with
the fewest cases for that surgeon). Or, put another way, if the Hosp
changes for a surgeon the indicator var "keepers" goes from a 0 to 1 and
stays that way until the next surgeon.
I realize there are a lot of programs out there (SSC). How does one go
about searching them for a possible solution to programming challenges?
*
* 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/
