I was just summarising the thread before adding a comment. I also added
a reference to a relevant paper.
My point about data structure (some say form, some say format) was
precisely the point made by Martin Weiss and Kit Baum, who gave
references to other material.
For example, the FAQ cited by Martin shows that there is a one-line
solution to your problem with a long data structure which is even
simpler than the loop several people suggested for your existing wide
data structure. Did you read it?
Nick
[email protected]
Polis, Chelsea B.
Nick Cox notes: "Chelsea's problem is easily soluble with the data
structure Chelsea has,
but a different data structure is advisable nevertheless."
Can you please clarify why a different data structure is advisable?
*
* 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/
