The whole issue is perhaps made clearer
by considering what happens when you -sort-
a dataset on a variable with some missing values.
In essence, the missing values have to go
somewhere, and being in the middle is out of the
question. So the choices are below the lowest (SAS)
or above the highest (Stata).
Yet another way to indicate "above some value,
but not missing" is to use (e.g.)
if inrange(2,.)
which means >= 2 but not missing.
Nick
[email protected]
evan roberts
> By way of making the point that this is a feature--not a bug--SAS
> considers all missing values to be less than the negative numbers. As
> with Stata it is considerably easier for programming purposes that
> logic be consistently two-way, provided you know what your missing
> data codes or values are.
*
* 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/
