My reply to Radu Ban
> I am going to interpret this as two Stata variables
> called -what- and -value-.
>
> First, we pull out the Group identifiers.
> One way to select them is that the previous
> value of -what- is missing. (This holds for the
> very first as well.)
The reasoning here is loose. This condition also applies
to other observations as well, but for those
the Group is blank, so that doesn't matter,
I think.
What's more sensitive is that the procedure
does assume some blanks separating each
pair of subgroups, i.e. that the number in each
subgroup is at least 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/
