None of the pundits on categorical data analysis have come forward so far,
so let me at least try to answer the first question: Are you sure you are
not mixing up the syntax of -predict- and -mfx- after -asclogit-? A look at
the respective ados shows that -mfx- takes integers with default 1 for -k-
while -predict- takes -string-s for -k-, i.e. can process "observed". BTW, I
also had to take a second look and switch the -trace- on to see the
difference. It is all too easy to overlook it in the -h asclogit post-...
HTH
Martin
*
* 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/
